Habitat and Species Cooperative Restoration Program Joint Base Lewis-McChord Center for Natural Lands Management

JBLM Lark Monitoring Final Report 2016 W911S8-15-2-0001 W911S8-15-2-0004 W911S8-15-2-0012 W911S8-16-2-0010 CNLM Task Orders #G1117, G1118, G1131, G1155 March 2017

Submitted to: Joint Base Lewis-McChord Fish and Wildlife Program

Submitted by: Adrian Wolf, Gary Slater and Jerrmaine Treadwell Center for Natural Lands Management 120 Union Avenue Southeast Olympia WA, 98501 Phone: 360-584-2538

Joint Base Lewis-McChord is a key military installation and the most important conservation area in the Puget Trough region. The Center for Natural Lands Management strives to assist Joint Base Lewis-McChord in the conservation of its natural resources within the framework of the military training mandate. Joint Base Lewis- McChord and its conservation partners have shared interests because:

 Healthy natural ecosystems are essential for realistic and sustainable training lands.  Rare species recovery throughout the region reduces the burden of recovery on any single landowner or site.  Pest plants harm natural areas and reduce their suitability for military training. Page i

Table of Contents Project Highlights ...... 1 1.0 Introduction ...... 2 1.1 Goals and Objectives ...... 2 2.0 Abundance and Distribution ...... 4 2.1 Methods ...... 4 2.1.1 Abundance Transects ...... 4 2.1.2 Surveying Sites with Unknown Occupation Status (Occupancy Surveys)...... 4 2.2 Results and Discussion ...... 5 2.2.1 Abundance Transects ...... 5 2.2.2 Occupancy Surveys ...... 5 3.0 Territory Mapping ...... 8 3.1 Methods ...... 9 3.1.1 Home Range ...... 9 3.1.2 General Use Area/Site Level ...... 10 3.1.3 Individual Use Area/Territory Level ...... 11 3.2 Results ...... 11 3.2.1 Home Range ...... 11 3.2.2 General Use Area/Site Level ...... 11 3.2.3 Individual Use Area/Territory Level ...... 12 4.0 Nest Monitoring ...... 16 4.1 Methods ...... 16 4.1.1 Number of Breeding Pairs ...... 16 4.1.2 Nests ...... 16 4.1.3 Clutch Initiation Date ...... 17 4.1.4 Vital Rates ...... 18 4.1.5 Characterizing Human Activities ...... 19 4.1.6 Communications with JBLM Airfield and Training Area Managers ...... 19 4.2 Results and Discussion ...... 20 4.2.1 Number of Breeding Pairs ...... 20 4.2.2 Nests ...... 20 4.2.3 Clutch Initiation Date ...... 21 4.2.4 Vital Rates ...... 22 4.2.5 Characterizing Human Activities ...... 25 4.2.6 Noteworthy Species ...... 27 5.0 Genetic Rescue at 13th Division Prairie ...... 27 6.0 Juvenile and Adult Return Rates ...... 29 6.1 Methods ...... 29 6.1.1 Resight Data ...... 29 6.1.2 Return Rates ...... 30 6.2 Results and Discussion ...... 30 6.2.1 Resight Data ...... 30 6.2.2 Return Rates ...... 30

SHLA_JBLM_2016_Final_2March_2017 Page ii

7.0 Optimizing Site Co-Use ...... 32 7.1 Breeding Phenology and Fledgling Vulnerability ...... 33 7.2 Communications with JBLM Airfield and Training Area Managers ...... 34 7.2 Fledgling Flush Distance ...... 36 8.0 Habitat Management at 13th Division Prairie ...... 38 9.0 Population Management Recommendations ...... 39 9.1 Coordinate Management Actions Closely with Nest Status ...... 39 9.2 Continue Lark Monitoring at Occupied Sites ...... 40 9.2.1 Territory mapping ...... 40 9.2.2 IButtons ...... 40 9.2.3 Occupancy surveys in new occupied sites ...... 41 9.2.4 Target trap adults ...... 41 9.3 Identify Nest Predators ...... 41 9.4 Convert Vegetation at Airfields ...... 41 9.5 Evaluate Effect of Disturbances ...... 42 9.6 Survey Other Suspected and Potential Lark Locations ...... 42 10.0 Acknowledgments ...... 43 11.0 Literature Cited ...... 44 12.0 Appendices ...... 48

List of Figures Figure 1. Streaked Horned Lark study sites on, and in the vicinity of, JBLM, WA in 2016...... 3 Figure 2. Maximum number of Streaked Horned Lark males and total maximum number counted on abundance transects surveys at five sites on JBLM, WA, 2010-2016 ...... 6 Figure 3. Streaked Horned Lark detection locations from abundance transects, 13th Division Prairie (top left), Gray Army Airfield, McChord Airfield, Range 76 and Range 50, in 2016 at JBLM, WA ...... 7 Figure 4. Streaked Horned Lark detection locations from occupancy transects, Range 50/52, and south McChord Airfield, JBLM, WA, 2016 ...... 8 Figure 5. General use areas, individual home ranges and nest locations for Streaked Horned Larks at 13th Division Prairie, JBLM, WA 2016 ...... 13 Figure 6. General use areas, individual home ranges and nest locations for Streaked Horned Larks at McChord Airfield, JBLM, WA 2016 ...... 14 Figure 7. General use areas, individual home ranges and nest locations for Streaked Horned Larks at Gray Army Airfield, JBLM, WA 2016 ...... 15 Figure 8. Number of nests initiated by date for Streaked Horned Lark nests at three sites, 2016 ...... 21 Figure 9. Causes of Streaked Horned Larks nest failure at three sites on JBLM, 2016...... 23 Figure 10: Nest outcome by nest host at 13th Division Prairie, Gray Army Airfield, McChord Airfield ..... 24 Figure 11: Human activities at 13th Division Prairie, 2013-2016 ...... 26 Figure 12: Signage and maps in occupied lark habitat at 13th Division Prairie in 2016 ...... 26 Figure 13. Observed paratrooper drop zone and staging areas, 13th Division Prairie, 2016...... 35 Figure 14: Distance, and age, of lark fledglings to observers when flushed, 2015-2016 ...... 38

SHLA_JBLM_2016_Final_2March_2017 Page iii

List of Tables Table 1. Maximum number of male, female and total Streaked Horned Larks counted on abundance transects surveys at five sites on JBLM, WA 2016 ...... 6 Table 2: Size of Streaked Horned Lark general use area at three sites, JBLM, 2015 and 2016. Volume contours calculated from kernel estimates (95% isopleth) with all location data...... 12 Table 3: Size of Streaked Horned Lark home ranges using kernel estimates (95% isopleth) for 16 individual use areas at three sites with data collected from April through 30 June 2016...... 12 Table 4. Estimated number of Streaked Horned Lark breeding pairs at seven sites on JBLM, 2016, 2015, 2014, 2013, 2003 and 2002...... 20 Table 5. Breeding summary of Streaked Horned Larks at three sites on JBLM, 2016...... 24 Table 6. Number of adult larks resighted and return rates, JBLM 2011 - 2016...... 32 Table 7. Number of color-banded lark nestlings that successfully fledged and their return rate after their first winter, JBLM 2010 - 2016...... 32 Table 8. Number of color-banded lark nestlings that successfully fledged and their return rate by sex after their first winter at three sites, JBLM 2015-2016...... 32 Table 9. South Puget Sound Streaked Horned Lark life stages and vulnerability ...... 33

List of Appendices Appendix 1. Streaked Horned Lark abundance transects surveys results at five sites, JBLM, 2016 ...... 48 Appendix 2. Color-banded Streaked Horned Lark adults, South Puget Sound, 2016 ...... 49

SHLA_JBLM_2016_Final_2March_2017 Page 1

Project Highlights

Population Size  Estimated Streaked Horned Lark breeding population on JBLM in 2016 as 87-92 pairs, representing a 61-64% increase relative to 2015 Abundance Surveys at Occupied Sites  Continued standardized lark monitoring at occupied sites  Surveyed 13th Division Prairie, Gray Army Airfield, McChord Airfield, Range 76 and Range 50, three times during the breeding season  Found that total number of lark detections across the five sites increased 18.5% in 2016, relative to 2015  Conducted occupancy surveys at fourteen sites with unknown occupation status  Documented new lark locations at three sites: the south field of McChord Airfield, Range 50 and Range 52 of the Artillery Impact Area Territory Sizes and Home Range  Calculated mean home range size for 16 individual territories as 11.6 acres (± 1.2 SE; Range: 4.0 - 20.6 acres) Nest Monitoring at 13th Division Prairie, McChord Airfield and Gray Army Airfield  Completed 16 weeks of nest monitoring surveys of 61-66 pairs  Located and monitored fate of 119 nests and color-banded 208 young of the year  Calculated traditional nesting success estimate of 66%; traditional nest success does not correct for failed nests that were not found  Calculated overall hatch rate of 89%; hatch rate on airfields was >90%  Determined overall number of fledglings per nest was 1.9 (± 0.1 SE) Adult and Juvenile Return Rates  Detected 68 banded adults in our study areas: 45 (66%) males and 23 (34%) females  Determined that 62.9% of the banded population were first-year breeding birds  Computed annual juvenile return rate in 2016 as 0.39 (0.53 at Gray Army Airfield, 0.36 at McChord Airfield, and 0.14 at 13th Division Prairie)  Calculated overall annual juvenile return rate since 2011 was 0.25  Determined that 10 of 43 (23.2%) first-year breeding larks dispersed to non-natal sites Optimizing Site Co-Use  Calculated mean distance that post-fledging larks flushed from observers was 11.5 m (± 0.7 SE, n = 144)  Found no known nests (eggs or nestlings) directly impacted by human activities in 2016.

SHLA_JBLM_2016_Final_2March_2017 Page 2

1.0 Introduction The Streaked Horned Lark (Eremophila alpestris strigata) is one of 24 genetically distinct subspecies of the Horned Lark that historically ranged from the Rogue and Willamette Valleys of Oregon, north to the Georgia Basin of British Columbia and coastal beaches in Washington (Drovetski et al. 2005, Stinson 2005). The current distribution of breeding Streaked Horned Larks has been reduced to the Willamette Valley, dredged-material islands of the Columbia River, coastal beaches of Washington, and South Puget Sound grasslands. In October 2013, the United States Fish and Wildlife Service listed the Streaked Horned Lark as threatened under the Federal Endangered Species Act (50 CFR part 17). The current population of Streaked Horned Lark in the Puget Sound region is an estimated 252 individuals (197-344 95% CRI, Keren and Pearson 2016). The majority of Puget Sound larks occur on Joint Base Lewis-McChord (JBLM) military base, making the tracking and enhancement of this important population and the improvement of habitat at occupied and unoccupied sites critical to its persistence in the region. Although recent estimates show a stable or increasing JBLM population (Keren and Pearson 2016), previous modeling indicated larks were declining and that actions to increase all three vital rates (fecundity, adult and juvenile survival) are necessary to stabilize the population (Camfield et al. 2011). Since 2011, The Center for Natural Lands Management (CNLM) has partnered with JBLM to assist in lark conservation and recovery through a variety of activities, including research, monitoring, and on-the-ground habitat management. In 2016, we continued several actions to support Streaked Horned Lark populations including occupied site monitoring (abundance and nest monitoring), communication with site managers, population enhancement (genetic rescue), habitat management, and research to understand juvenile and adult survivorship. 1.1 Goals and Objectives The goal of our work is to conduct conservation actions on JBLM aimed at stabilizing or reversing the observed decline in Streaked Horned Lark populations. We conducted work at all known occupied sites on JBLM in the Puget Lowlands of Washington State (Fig. 1). The objectives of our work were nine-fold, and correspond to subsequent sections of this report: 1) Contribute data to assist Washington Department of Fish and Wildlife (WDFW) in estimating abundance and trends through time at five occupied sites (13th Division Prairie, Gray Army Airfield, McChord Airfield, Range 76 and Range 50) (Section 2.0); 2) Conduct occupancy surveys at sites that support potentially suitable lark habitat but where lark status is unknown (Close-In Training Area F, South McChord Airfield, Artillery Impact Area, and Training Areas 6, 13, 14, 15, 21, 22, and 23) (Section 2.0); 3) Collect spatial data to better understand breeding season home range (Section 3.0); 4) Collect information on breeding status and reproduction at three sites (13th Division Prairie, Gray Army Airfield, McChord Airfield) (Section 4.0); 5) Continue to implement and monitor results of the genetic rescue study (Section 5.0); 6) Collect information to minimize, risks to juvenile larks (Sections 6.0 and 7.0); 7) Continue research to understand survivorship at multiple life stages (Section 6.0); 8) Implement habitat enhancement activities to improve foraging and nesting habitat. 9) Outline management recommendations that would provide for the increase in, and recovery of, the JBLM Streaked Horned Lark population.

SHLA_JBLM_2016_Final_2March_2017 Page 3

Figure 1. Streaked Horned Lark study sites on, and in the vicinity of, JBLM, WA in 2016.

SHLA_JBLM_2016_Final_2March_2017 Page 4

2.0 Abundance and Distribution Understanding the abundance and spatial distribution of Streaked Horned Larks at JBLM is essential to direct habitat management, balance multiple uses (e.g., military training, lark breeding), and monitor population trends. Since 2010, Streaked Horned Larks have been the target of a consistent monitoring effort on occupied sites at JBLM using a standardized monitoring scheme developed by WDFW and implemented across all known occupied sites in Washington (Pearson et. al., 2016). In 2016, we continued surveys of larks at JBLM occupied sites (i.e., 13th Division Prairie, Gray Army Airfield, McChord Airfield, Range 74/76, and Range 50) and surveys of unoccupied, yet potentially suitable sites (e.g., TA6). 2.1 Methods We performed two types of surveys with different objectives:

2.1.1 Abundance Transects We followed WDFW’s standardized protocol for assessing abundance and trends in Washington at five occupied sites on JBLM (Pearson et. al., 2016):  13th Division Prairie – 13,504 m total transect length  Gray Army Airfield – 5,995 m total transect length  McChord Airfield – 9,614 m total transect length  Range 76 – 5,400 m total transect length  Range 50 – 5,398 m total transect length (increased survey area relative to 2015)

We extended the Range 50 survey routes to the west because of an early season detection of larks in this area (see Section 2.2.2). In general, surveys were conducted between 30 minutes after sunrise and noon on days when wind was < 20 mph with little to no precipitation (light drizzle and brief showers were acceptable). Surveys ended by 11:00 a.m. or earlier on days when the temperature was 80° F or higher. From 2010 to 2013, sampling was conducted three times during the lark breeding season - once in May, June and July along predetermined line transects spaced 150 m apart and placed to cover the majority of potentially suitable habitat at each site. In 2014, following guidance from WDFW, the sampling protocol was modified so that all three surveys at each site were conducted within a six-week period between May and mid-June. Observers walked at a pace of 1-2 miles per hour. Observers mapped and tabulated all larks seen or heard and recorded initial detection type (aural vs. visual); observers also recorded vocalization type (song or call), sex and age if known, secondary behaviors, environmental variables, and predator/competitor tallies.

2.1.2 Surveying Sites with Unknown Occupation Status (Occupancy Surveys) We performed two types of occupancy surveys designed to assess whether areas were occupied by larks. The first type consisted of survey routes, spaced 150 m apart, followed the protocol of Pearson et. al. (2016), and were conducted at two sites:  Close-In Training Area F (CTAF,) – 2,476 m total length  South McChord Airfield – 7,815 m total length

SHLA_JBLM_2016_Final_2March_2017 Page 5

We also conducted modified occupancy surveys along 110 line transects (each 250m in length) in suitable lark habitat on four prairies where lark status was unknown: 30 transects in 13th Division Prairie (Training Areas 13-15); 27 transects between the Weirs and Johnson Prairies (Training Areas 21-23); and 53 transects in the Artillery Impact Area of 91st Division Prairie (Ranges 50, 52/53, 74/76; Training Area 6; and Mortar Points 1-3 and 8-13) (Fig. 1). Line transects were typically completed in 15-20 minutes, and all bird species <150 m from the line were recorded and their perpendicular distance from the transect line was estimated. In total, we conducted 327 line transects during three general survey periods: 2-18 May (110 transects), 9 May -20 June (110 transects), and 27 May –27 June (107 transects). 2.2 Results and Discussion

2.2.1 Abundance Transects In 2016, we conducted three repeated surveys at all five JBLM occupied sites. Over the 15 surveys, we recorded 283 lark detections (Table 1, Figs. 2-3, Appendix 1), although many of these detections were assumed to be repeat detections of the same bird at a given site. The highest individual count occurred at Gray Army Airfield with 42 larks (28 males); the other four sites had high counts of 15-25 larks (Appendix 1). Total maximum detections across the five sites increased 18.5% in 2016, relative to 2015 (Fig. 2). Three sites showed the greatest increase: Gray Army Airfield, and the two sites in the Artillery Impact Area (Range 50 and Range 76), whereas 13th Division Prairie remained stable. We have used the high counts of male larks detected during any one survey as an estimate of the number of breeding pairs. Because we were simultaneously monitoring nesting pairs and conducting abundance transects, we documented that this assumption underestimated the presumed breeding population in 2016 (Table 1). Based on maximum counts of males, the abundance data indicated 55 pairs occurred on the three sites (13th Division, Gray Army Airfield and McChord Airfield), which underestimated the presumed breeding population estimated from nest monitoring (61-66 pairs). The maximum counts of males provided a close estimate of the number of pairs at Gray Army Airfield and 13th Division Prairie, but underestimated the number of pairs, by 5-10, at McChord Airfield (Table 1).

2.2.2 Occupancy Surveys The occupancy transects documented new lark locations at three surveyed areas: the south field of McChord Airfield, Range 52, and the western portion of Range 50 of the Artillery Impact Area (Fig. 4). The larks detected in the south field of McChord Airfield were a color- banded breeding pair, which represented a new breeding location for the airfield. The second new location entailed the detections of two male larks in the vicinity of Range 52. As a result of this occurrence, we recommend that this area of the AIA be surveyed in early 2017 to assess occupancy. If larks are determined to be present, we recommend conducting abundance transects in this vicinity. The third new location comprised the detection of 6 larks (unknown sex) in the western portion of Range 50. As a result of the new detections of larks in the western portion of R50 in early May, we expanded the abundance transect routes to encompass this area. Despite the extension of abundance transects survey routes to the west, we did not detect larks in this area during the abundance transect surveys, and only recorded larks during the occupancy surveys. Habitat

SHLA_JBLM_2016_Final_2March_2017 Page 6

appears suitable at Range 52, Range 53, Training Area 6, Training Area 14, Training Area 15 and Upper Weir Prairie (Training Area 21). We detected a total of 3,257 individuals, representing 73 species on the occupancy line transects. The four most abundant species (average number of individuals per transect + S.D.) were Savannah Sparrow (Passerculus sandwichensis; 4.3 + 1.2), White-crowned Sparrow (1.0 + 0.1), Chipping Sparrow (Spizella passerina; 0.7 + 0.2), and Western Meadowlark (Strunella neglecta; 0.5 + 0.1).

Table 1. Maximum number of male, female and total Streaked Horned Larks counted on abundance transects surveys at five sites on JBLM, WA 2016

Site Max #detections+ Max #Males+ Max# Females+ Number of Breeding Pairs* 13th Division Prairie 15 12 1 11 (+ 2 unpaired males) Gray Army Airfield 42 28 10 30 (+ 2 unpaired males) McChord Airfield 25 15 3 20-25 Range 50 15 9 0 9 Range 76 15 12 3 12 Total 112 76 17 82 - 87 + maximum number on a single survey; *# pairs determined by nest monitoring, or max number of males at R50 and R76

30 13DIV GAAF MAFB

R76 R50 25

20

15

10

Max. number of males 5

0 2010 2011 2012 2013 2014 2015 2016 50

40

30

20 detections

Totalnumber of 10

0 2010 2011 2012 2013 2014 2015 2016

Figure 2. Maximum number of Streaked Horned Lark males (top) and total maximum number (bottom) counted on abundance transects surveys at five sites on JBLM, WA, 2010-2016.

SHLA_JBLM_2016_Final_2March_2017 Page 7

Figure 3. Streaked Horned Lark detection locations from abundance transects, 13th Division Prairie (top left), Gray Army Airfield (top center), McChord Airfield (top right), Range 76 (bottom left) and Range 50 (bottom right), in 2016 at JBLM, WA.

SHLA_JBLM_2016_Final_2March_2017 Page 8

Figure 4. Streaked Horned Lark detection locations from occupancy transects, Range 50/52 (left), and south McChord Airfield (right), JBLM, WA, 2016.

3.0 Territory Mapping Animal space-use models assume that individuals space themselves and move through their environments in ways that increase fitness via the efficient exploitation of resources and avoidance of risks (Spencer 2012). Therefore, location data of individuals should not be considered random occurrences, but rather as critical information essential for the species ecological needs. Each location could be functioning as an important behavioral, dispersal and/or breeding venue. Generating spatially explicit home range maps identify these critical locations for breeding larks, and these maps could be used to assist JBLM managers with training planning and habitat management decisions, while addressing the need to minimize impacts to larks. These data could help us also understand how larks respond to habitat manipulation, by monitoring the size and shape of home ranges, overlap among individuals, movement patterns within home ranges, and boundaries changes over time. The goal of the territory mapping effort was to characterize lark use of three occupied sites: 13th Division Prairie, McChord Airfield and Gray Army Airfield. The objectives of the territory mapping activities were four-fold: 1) delineate overall site use-areas; 2) calculate

SHLA_JBLM_2016_Final_2March_2017 Page 9

breeding home range size; 3) understand annual changes in lark use area among years; and 4) evaluate lark response to habitat enhancement activities. For this analysis, we did not follow the strict definition of territory as “any defended area” (Noble 1939). Rather, we used the term territory interchangeably with the concept of an individual’s or pair’s home range or use area (Odum and Kuenzler 1955). The definition of a home range is an area routinely used by an animal to meet its daily needs (Odum and Kuenzler 1955), and is less restrictive than territory. We believe this approach is appropriate because our goal was to capture the full spectrum of habitat conditions that larks use during the breeding season. Moreover, the lark is not as territorial as most passerines and often tolerates the close presence of adjacent breeding individuals, making it difficult to determine the defended space boundary for an individual or pair.

3.1 Methods

3.1.1 Home Range We collected home range information on three of our study sites: 13th Division Prairie, McChord Airfield, and Gray Army Airfield. The goal of the territory-level mapping effort was to obtain at least 15 territory points for each use area/territory on the three study sites in the early to mid-breeding season (early-April to 30 June). To characterize general use areas at each site, we continued to collect territory points through the late breeding season (August), and pooled these data with the territory-level data. The protocol for the territory mapping efforts was as follows: we identified general lark distribution during transect surveys (sections 2.0, 6.1.1) and nest monitoring activities (section 4.0) and used this information to guide and focus our territory mapping effort. Upon locating an adult lark (female or male), observers followed the resighting protocol (section 6.1.1) to ascertain whether the bird was color-banded. We recorded spatial data for each pair/use area where at least one adult’s color-marked status was confirmed. If male and female individuals of a pair were both unmarked, we recorded territory information for those individuals only if they could be differentiated from marked individuals of the same sex in adjacent use areas. For example, if two unbanded females shared a common use-area boundary, we did not collect spatial data for either female. After band-status was determined, observers spent 30 minutes following the bird to obtain 5-7 territory points as the individual moved through its territory. In addition to these focused territory mapping sessions, we also recorded spatial data opportunistically during nest monitoring activities (see Section 4.0). Observers collected location data (UTM coordinates) with handheld GPS units, and recorded time, and behaviors (e.g., singing, agonistic, foraging, etc.).

SHLA_JBLM_2016_Final_2March_2017 Page 10

Home Range Analysis There are numerous techniques for estimating home ranges by animals and each has their strengths and weaknesses. The minimum convex polygon (MCP) method has traditionally been the most commonly used approach, and the method simply connects the outermost points in a location date set (Mohr 1947). This method is highly affected by locations on the periphery, and it can contain large areas that were never visited by the focal organism (Harris et al. 1990). More recently, two methods - kernel density estimate (KDE) and local convex hull (LoCoH) – have been developed to address problems with MCP. These methods use computational techniques to create a utilization distribution (UD) in 2-dimensional space based on the density of points, which allows UD to be calculated for different isopleths (e.g., 50%, 95%). Both methods require user-defined tuning parameters and are more sensitive to independence of points. At its simplest, KDE can be thought of as creating a bump or buffer around points, whereas LoCoH creates minimum convex polygons around the three closest points (hulls) and combines these to estimate a density distribution. One advantage of LoCoH is that it better resolves narrow use-corridors and sharp boundaries, such as those produced by habitat edges, including roads (Getz and Wilmers 2004). In 2015, we evaluated these three methods for estimating lark home ranges using location data from 13th Division Prairie with the goal of identifying the most biologically meaningful results (Wolf et al. 2016). Results indicated home ranges estimate using the KDE method proved suitable for our objective. Further, we also investigated KDE estimates calculated from 1) all detections, 2) initial detections and 3) subsequent detections (>1 detection per day if time-lapse between detections ≥20 minutes) because subsequent detections during a given observation bout may not be considered independent (Hejl et. al. 1990). Results showed little difference between different sets of detections. Based on results from 2015, we continued to estimate home ranges with KDE using all location data in 2016. Bandwidths were generated for overall location data for each site and from use-area datasets where location sample data was ≥ 15. Bandwidths were calculated with the statistical software Program R (Version 3.2.3), and KDE was calculated with Geospatial Modeling Environment (Version 0.7.2.0, Beyer 2014). We generated 95% (home range areas) volume contours (isopleths) to capture 95% of the probability that larks could be located within two home range areas: 1) general use areas, and 2) individual use areas/territory level. We compared changes in general use-area size and distribution between 2015 and 2016.

3.1.2 General Use Area/Site Level A general use area was defined as the home range comprised of all breeding territories within a site. We generated general use area home range for each site using all territory mapping location data collected during the entire breeding season (April through August). The purpose of estimating general use area by site was to delineate and illustrate the overall distribution of larks within a site. A graphical illustration of site-level use by larks would help inform sites managers where larks occur, which could facilitate planning, training and management activities, while minimizing impacts to breeding birds. In addition, the general use areas would provide spatially explicit information of lark response to habitat enhancement activities.

SHLA_JBLM_2016_Final_2March_2017 Page 11

3.1.3 Individual Use Area/Territory Level An individual use area was defined as the home range used by a territorial male (paired or unpaired) between 1 April and 30 June. We generated individual use areas by pair to 1) estimate and compare home range sizes for lark breeding pairs at the three study sites; and 2) to understand lark use area responses to habitat enhancement activities at 13th Division Prairie.

3.2 Results

3.2.1 Home Range We recorded a total of 802 territory mapping locations (212 locations were paired birds, 397 locations were only males, and 193 were only females). Because we obtained more location data for males, our home range calculations might have overestimated overall site use and individual use areas. Males, as compared to females, typically use larger areas because they are frequently found displaying, guarding females, and defending territories. For example, males are more likely to be detected beyond their regular use areas after an agonistic event (e.g., chase). Female larks, the other hand, were more frequently detected within their defined territories, and generally near their nest sites, presumably because female larks conduct all nest building and incubation activities without the assistance of their mates.

3.2.2 General Use Area/Site Level The territory mapping effort showed that larks do not use the three sites uniformly (Figs. 5 - 7). The proportion of the total site mapped by the 95% isopleth for the general use area was 36% for 13th Division Prairie, 40% for McChord Airfield, and 75% for Gray Army Airfield) (Table 2). Possible explanations for the concentration of larks in these areas include the presence of suitable vegetation structure and composition (Pearson and Hopey 2005, Kronland et. al. 2016), strong site fidelity, and the propensity of larks to aggregate as semi- colonial breeding birds. The biggest change in site use between 2015 and 2016 was on 13th Division Prairie. Site use at 13th Division Prairie in 2016 was no longer clustered in three general areas, as it was in 2015 (Fig. 5). The difference between years was evident in the central portion of the study area (i.e., the Cobble Beds) which was used by two pairs, and at least one unpaired male. This area received considerable habitat work in late 2015, including the removal of Scotchbroom, blackberry, and several Douglas firs (see Section 8, and Kronland 2016). On the two airfields the general use area did not vary much from 2015. Site use at McChord Airfield in 2016 was similar to 2015, with the exception of an isolated area in the south field (Fig. 6). The general use area at McChord Airfield comprised 40% of the study area, and the south field location was occupied by a single breeding pair, which was detected late in the season during an occupancy survey. At Gray Army Airfield, the general use area continued to be evenly spread across the study area and comprised 75% of the site (Fig. 7).

SHLA_JBLM_2016_Final_2March_2017 Page 12

3.2.3 Individual Use Area/Territory Level We estimated home ranges for 16 pairs of larks on nesting territories at the three sites (10 at 13th Division, and 3 each at McChord Airfield and Gray Army Airfield, Table 3, Figs. 5-7). The KDE method mean home range size was 11.6 acres (± 1.2 SE), and ranged in size from 4.0 to 20.6 acres. Home ranges overlapped substantially, as suspected given the semi-colonial breeding behavior of the species. Further research is necessary to understand territory size differences among sites.

Table 2: Size of Streaked Horned Lark general use area at three sites, JBLM, 2015 and 2016. Volume contours calculated from kernel estimates (95% isopleth) with all location data.

2015 2016 Site* Acres (% of study area, n) Acres (% of study area, n) 13th Division Prairie 165.9 (30%, 390) 202.7 (36%, 335) Gray Army Airfield 266.6 (76%, 182) 266.5 (75%, 272) McChord Airfield 251.9 (34%, 420) 299.6 (40%, 317) All Sites 684.4 (41%, 992) 768.8 (47%, 924)

Table 3: Size of Streaked Horned Lark home ranges using kernel estimates (95% isopleth) for 16 individual use areas at three sites at JBLM, 2016.

Home Range (95% Isopleth) Site Use Area# N Acres 01 18 4.04

13th Division Prairie 02 25 8.57 03 26 8.51

04 23 9.91

05 40 10.66 07 31 14.34

08 23 13.48

09 17 7.68 11 32 13.67

13 27 20.57

Ave ± SE 11.14 ± 1.45

01 20 13.29 McChord Airfield 02 23 11.60 10 18 20.57

Ave ± SE 15.15 ± 2.75

14 21 10.80

Gray Army Airfield 17 16 14.31 26 22 4.26 Ave ± SE 9.79 ± 2.94 All Sites Ave ± SE 11.64 ± 1.18

SHLA_JBLM_2016_Final_2March_2017 Page 13

Figure 5. General (right) and individual (left) use area home ranges and nest locations for Streaked Horned Larks at 13th Division Prairie, JBLM, WA 2016.

SHLA_JBLM_2016_Final_2March_2017 Page 14

Figure 6. General (right) and individual (left) use area home ranges and nest locations for Streaked Horned Larks at McChord Airfield, JBLM, WA 2016.

SHLA_JBLM_2016_Final_2March_2017 Page 15

Figure 7. General (right) and individual (left) use area home ranges and nest locations for Streaked Horned Larks at Gray Army Airfield, JBLM, WA 2016.

SHLA_JBLM_2016_Final_2March_2017 Page 16

4.0 Nest Monitoring Monitoring Streaked Horned Lark breeding activity and their nests is the most accurate and efficient method to estimate reproductive success, identify factors associated with fecundity, and direct management at occupied sites to minimize negative impact to breeding larks. The goal of the nest monitoring effort was to monitor lark breeding at three sites: 13th Division Prairie, McChord Airfield, and Gray Army Airfield (Fig. 1). The objectives of the nest monitoring activities included: 1) estimating the number of lark breeding pairs, 2) identifying locations of lark territories and nests to inform management, and 3) collecting information on breeding status, nesting success, and productivity. 4.1 Methods

4.1.1 Number of Breeding Pairs The overall estimated number of breeding pairs on JBLM were calculated from the following data: 1) known number of nesting pairs determined through nest monitoring activities; 2) maximum number of males (assumed to be paired) detected during the abundance and occupancy transects at Range 50, Range 52, and Range 76 (section 2.0); and 3) the conservative assumption that half of the six larks detected during the occupancy transects in Range 50 during one visit, were male and paired (section 2.2.2). At the three nest monitored sites (13th Division Prairie, Gray Army Airfield and McChord Airfield), we estimated the number of breeding pairs by analyzing spatial data from the transect surveys (sections 2.0, 6.1.1), territory mapping activities (section 3.0), and resight surveys (section 6.0). We used these data to supplement our nest monitoring observations of marked and unmarked adults and their nest locations, and plotted the registrations and locations of breeding pairs on aerial maps (scale 1 inch = ~250m). Where two or more pairs were closely packed, we used male territorial displays and behaviors (e.g., singing, agnostic behavior, mate guarding) to distinguish breeding pair boundaries (Bibby et al. 1992).

4.1.2 Nests We identified general lark distribution during transect surveys (sections 2.0, 6.0), territory mapping (section 3.0), and juvenile resight surveys (section 6.0) and used this information to guide and focus our nest monitoring activities. Nest locating and monitoring methods were adapted from Martin and Geupel (1993). We focused behavioral observations primarily on females because male larks neither assist with nest building nor incubation. Female larks were located and then followed with spotting scope or binoculars for signs of nest building or nest tending activity. Female larks excavate a depression, and then line the excavation with plant material and mosses, and may add rocks and pebbles on the north side of the nest (Beason 1995). When conspicuous nest building activity was not observed and the female was presumed to be incubating, areas where female activity was concentrated were searched on foot. Because male larks tend nestlings and fledglings, we only followed males when they were observed carrying food items. To reduce confusion about the status of nesting activities or how data were used in analyses, we defined nest status in the following way. We defined a nest as a case where >1

SHLA_JBLM_2016_Final_2March_2017 Page 17

egg was laid in a nest structure (Dinsmore et al. 2002). Data from these nests were used for all demographic analyses, including clutch size, nest success, and productivity. However, we also recorded the locations of all nest structures during the nest-building stage and provided that information to managers as a potential nest site. In some cases these sites eventually became nests, but in some cases they did not. Later in the season, we also found nest structures that had been nests earlier in the season based on evidence surrounding the nest structures (e.g., flattened nest structure, presence of fecal matter). We kept a location record of all these unfound nests, but these were not included in data analyses because nest outcomes were unknown. Finally, in some cases, we observed adults tending dependent young in lark territories where no nest had been found. For these cases, we recorded a nest location at the center of the delineated territory (section 3.0). These cases were not used in demographic analyses, but were considered when estimating overall young produced by site. Once a nest was discovered, we visited it every 3-5 days, until it fledged or failed. Nests were checked more frequently near fledging to accurately assess fate. We used a variety of cues to assess whether a nest was successful or failed. We assumed a nest was successful if young were > 6 d old at the previous visit and there were signs associated with fledging (adults alarm calling or feeding young, fecal matter on rim of cup, feather scales) and no signs of predation (feathers, disturbed nest cup). We banded lark nestlings to track translocated individuals and estimate return rates of adults and juveniles (Section 6.0). Most lark nestlings were banded at 6-8 days old. Each individual received an acetal or darvic color band over a silver metal USFWS band on the right leg. To differentiate among sites, birds at 13th Division Prairie received a yellow color band over metal; birds at McChord Airfield received dark blue over metal; and birds at Gray Army Airfield were given light blue or red over metal. Unique color band combinations were also placed on the left leg to differentiate among individuals. One central retrice was taken from each nestling for genetic analysis and stored at the WDFW. Unhatched eggs were removed after nests were abandoned or nestlings had fledged, and eggs with embryos were delivered to Scott Pearson (WDFW) for genetic analyses. All birds were banded under Bird Banding Permit #22913 or #22932. Nest Host Female larks generally build nests on the north or northeastern side of a vegetation clump (i.e., nest host). At each nest, we identified to species the plant or structure that served as the nest host.

4.1.3 Clutch Initiation Date When we knew nest age, we estimated clutch initiation dates by backdating using the following formula: egg laying stage (# eggs laid-1), incubation period (12 days), and nestling period (9 days). Otherwise, we used this formula to estimate the clutch initiation date (Pearson 2003): Clutch initiation = date found – (incubation period (12) – number of days observed) ÷ 2.

SHLA_JBLM_2016_Final_2March_2017 Page 18

4.1.4 Vital Rates Hatch Rate To assess any changes in hatchability following the genetic rescue effort (section 5.0), we calculated hatch rate at each site as the average number of hatched eggs by the total number of eggs laid by nest. We considered only those nests with completed clutches that survived the incubation period. Furthermore, manipulated nests from the genetic rescue effort (e.g., translocated/augmented clutches) were excluded from this analysis. Nest Success We used the traditional method of estimating nest success by dividing the number of successful nest attempts by the total number of nests known at each site. This procedure generally overestimates nest survival because it does not take into consideration nests that failed before they were found. In 2017 (the fifth year of consistent monitoring at all five sites), we intend to examine multiple covariates to investigate several factors that might have the biggest effect on daily nest survival using Program MARK (White and Burnham 1999), which yields an unbiased estimate of nest survival.

IButtons We deployed Thermochron iButtons (Model DS1921G) in 12 nests as a pilot study to increase efficiency in assessing nest outcomes (Potterf 2017). iButtons are temperature data loggers that have been used to monitor avian nesting success, offering a low cost and low effort alternative to typical nest monitoring techniques. We first tested the feasibility of iButtons on eight nests of abundant ground-nesting prairie species (Savannah Sparrow, Western Meadowlark, White-crowned Sparrow) to understand whether the technology influenced parental behavior and accurately indicated nest status, including predicting hatch and completion dates. Because we found that iButton placement did not lead to higher rates of abandonment with the lark surrogates, we then proceeded to test iButtons in four lark nests at 13th Division Prairie and McChord Airfield. Thermochron iButtons were set to collect temperature at 15-minute intervals and remained in the nest until nest completion. Target nests received one of three treatments: 1) one iButton was placed beneath the nest contents; 2) nest was physically manipulated in the same manner as placing an iButton without deploying the technology; and 3) nest was approached, but not manipulated (control). Immediately after each treatment, monitors collected behavioral observations for 45 minutes. For larks, we deployed iButtons in nests at the nestling phase to minimize rejection and abandonment by the parents (Boulton et al. 2009). Control and manipulated nests were visited every 3 or 4 days until completion.

Fledglings per nest To obtain a measure of fecundity (productivity), we estimated the mean number of fledglings per nest by site.

SHLA_JBLM_2016_Final_2March_2017 Page 19

4.1.5 Characterizing Human Activities We calculated the frequency of human activities on larks at 13th Division Prairie during our survey days. This analysis was not conducted at Gray Army Airfield or McChord Airfield because aircraft activities were daily occurrences and mowing operations were weekly activities early in the breeding season. For the assessment at 13th Division, we characterized observed activities into six categories (i.e., military vehicles, other vehicles, aircraft, recreation, paratroopers, and other) that might have resulted in direct lark mortality.

4.1.6 Communications with JBLM Airfield and Training Area Managers Results of nest monitoring activities were communicated to airfield and training area managers on at least a weekly basis. From 20 April through 2 September, we provided site managers with spatially explicit maps of each occupied site. The maps illustrated the location of potential nest sites (i.e., where nest building was observed), nests (active, fledged, failed) and sites where there was strong evidence of breeding based on behavioral cues. Maps also indicated recommended stay out areas (BANAs, or buffered active nesting areas) and vulnerability periods. Nests were assigned a vulnerable period date, defined as up to 14 days post fledging (total of 35 days). For example, a nest found on 1 June on day 1 of the egg-laying stage was assigned a vulnerable period until 5 July (3 days laying + 12 days of incubation + 9 days as nestlings + 14 days as fledglings). The decision to extend the vulnerability 14 days post –fledging is because during that time fledglings are weak flyers and may not be able to avoid military and management vehicles (e.g., mowers, see Section 7.2). JBLM managers also installed new signage at egress and ingress locations to priority habitat at 13th Division Prairie to limit unauthorized use (Fig. 12). All vehicular traffic (including lark monitoring crews) were directed to remain on gravel roads when training/operating within the perimeter of the priority habitat polygon. In addition, when active nests were located within 5 m of roads at 13th Division Prairie, signs were placed at strategic locations to indicate that the road was temporarily closed. Furthermore, training area managers identified a paratrooper exclusion zone/polygon to redirect training activities (particularly paratrooper drops and land maneuvers) from impacting critical nesting areas (Fig. 13).

SHLA_JBLM_2016_Final_2March_2017 Page 20

4.2 Results and Discussion

4.2.1 Number of Breeding Pairs In 2016, we estimated that 61-66 breeding Streaked Horned Lark pairs occurred at the three nest monitored sites (11 at 13th Division Prairie, 30 at Gray Army Airfield, and 20-25 at McChord Airfield, Table 4). This reflects a 38-39% increase relative to 2015, with the greatest increases at the two airfields. We recorded at least four unpaired males at two sites – 2 each at 13th Division Prairie and McChord Airfield. The overall JBLM breeding population was estimated to be 87-92 breeding pairs in 2016, which represents a 61-64% increase, relative to 2015.

Table 4. Estimated number of Streaked Horned Lark breeding pairs at seven sites on JBLM, 2016, 2015, 2014, 2013, 2003 and 2002.

Site* 2016 2015 2014 2013 20031 20022 13DIV 11 (+2 males) 9 (+2 males) 10 (+2 males) 9 (+1 male) 10 8 GAAF 30 19-21 (+2 males) 10 – 13 11 30 - MAFB 20-25 16-18 8 – 9 8 – 9 - - R763 12 6 8 - - - R50E3 9 3 9 - - - R50W4 3 - - - - - R525 2 - - - - - Total 87 - 92 53 - 57 45 - 50 27 - 30 >40 27 * Site codes: 13DIV = 13th Division Prairie, GAAF = Gray Army Airfield; MAFB = McChord Airfield; R50, R52 and R76 = Range 50, 52 and 76 of the Artillery Impact Area; 1 entire population estimated at 13DIV and GAAF, but only a portion of MAFB (Pearson and Hopey 2004); 2 entire population estimated at 13DIV (Pearson and Hopey 2004); 3 based on maximum number of males detected during a single abundance transect visit, 4 based on the assumption that half of the six individuals detected during one single occupancy transect visit were male and were paired, 5 based on the assumption that two males detected during occupancy transect were paired. Dash indicates no surveys were conducted.

4.2.2 Nests At all three sites in 2016, we found a total of 119 nests which contained 396 eggs and 227 nestlings (Table 5, Figs. 5-7). We color-banded 208 nestlings and fledglings between 21 April and 14 August 2016 (38 at 13th Division Prairie, 97 at Gray Army Airfield, and 73 at McChord Airfield). Not all nests were found, as evidenced by four instances where either a nest structure was found with evidence of breeding or we found adults tending dependent young in use areas where no nest was located.

Nest Host Although we have not quantitatively assessed the availability of nest host plants at each site, nest host selection appears to be non-random. Our nest monitoring efforts at 13th Division Prairie documented that 54.5% (12 of 22) of nests were built on the north side of the native bunchgrass, Festuca roemerii, and 32% of the nests (7 of 22) were built using English Plantain,

SHLA_JBLM_2016_Final_2March_2017 Page 21

Plantago lanceolata. The remaining nest hosts at 13th Division included the native forb, Lupinus lepidus (9.1%), and the rhizomatous grass, Agrostis capillaris (4.5%). The most commonly used nest host at Gray Army Airfield was English Plantain, where it accounted for slightly more than half of the nests (51.1%; 23 of 45), followed by the native forb Lupinus lepidus (13.3%, 6 of 45), and other grasses and non-native forbs (35.6%, e.g., Agrostis capillaris, Cytisus scoparius, Festuca roemerii). At McChord Airfield, grasses and non-native forbs (e.g., Centaurea pretense) comprised 55.3% (24 of 47) and 23.4% (9 of 47) of the nest hosts, respectively.

4.2.3 Clutch Initiation Date In 2016, the first known clutch was initiated in late March (~28 March), the earliest clutch initiation known to date. This date was estimated based on the detection of a fledgling that was color-banded on 21 April. We assume that the initiation of early laying was associated with the spell of warm weather in mid-to-late March. However, based on few observations of fledglings, it is likely that only a few pairs initiated early clutches. Most 2016 nests were initiated in three periods: mid-to-late-May, mid-June and mid-July (Fig. 8). Clutch size in 2016 was similar to 2015, with a mean of 3.4 (+ 0.1 SE; Table 5). Average clutch size was similar across sites. Of note were 59 nests where clutch size exceeded the typical, average, 3-egg clutch, indicating favorable environmental conditions for breeding. We documented 45 nest attempts with 4-egg and another 14 attempts with 5-egg clutches. The incidence of 5-egg clutches is rare for the species (Randy Moore, pers. comm.).

0.16

0.12

0.08

Percent Percent total of nests 0.04

0.00

Date Figure 8. Number of nests initiated by date (in 15 day intervals) for Streaked Horned Lark nests (n = 112) at three breeding locations in JBLM, 2016.

SHLA_JBLM_2016_Final_2March_2017 Page 22

4.2.4 Vital Rates Hatch Rate Hatch rate in 2016 was 89% at all three sites (Table 5), which is similar to rates reported for birds in general (>90%, Koenig et al. 1982). Hatch rates were ≥90% at both airfields, and less than 80% at 13th Division Prairie. Hatchability of Streaked Horned Lark eggs in the Puget Lowlands was reported as low as 44% in 2007 and 2009 (Anderson 2010). Because genetic factors (inbreeding depression) appear to be a likely explanation (Drovetski et al. 2005), a genetic rescue effort to increase genetic diversity was initiated in 2011 (see Section 5.0). Nest Success We calculated traditional nest success, defined as the percent of all nests found that successfully fledged at least one young, as 66% (79 of 119). Nest success was higher at 13th Division Prairie (77%), relative to the airfields (70% and 58% at Gray Army and McChord Airfields, respectively) (Table 5). However, these estimates should be interpreted with caution because they have not been corrected to account for nests that failed before they were found. Of the 42 nests that failed, the primary cause of nest failure was predation (n = 35). Other causes of nest failure included abandonment (n = 5), flooding (n = 2), and one nest where the eggs never hatched but the female continued to incubate (Fig. 9). Of the 35 predation events, we recorded 25 (60%) incidents where nest contents disappeared with no apparent sign of predation. The remaining 10 predation events were those with damaged nest contents (e.g., broken eggs, nest pulled out of the ground). Although we are not certain of the specific predators, we suspect that these nest-failure signs reflect different predator suites (Fig. 9). Our results indicate that overall predation rates are similar among sites, but the type of nest-failure sign indicates that predators may be different among sites. Wildlife management at McChord Airfield (APHIS) implement activities to dissuade and eliminate potentially hazardous wildlife (e.g., corvids and coyotes), but we do not understand whether these activities affect predation rates at lark nests. Nest cameras would help determine the specific causes of nest failure, and we intend to initiate a study in collaboration with JBLM beginning in 2017. For the second year in a row, we recorded no incidents where airfield or military training activities resulted in nest failure. We believe this is due to the improved communication of information regarding lark distribution and status to airfield managers. Although sample sizes are small, there appears to be differences in nest outcome when considering the particular plant species that were selected as nest hosts. For example, most nests hosted by the native bunchgrass, Festuca roemeri (88%, n = 17), and English Plantain, Plantago lanceolata (77%, n = 39), were successful. Conversely, rhizomatous grass and other non-native forb nest host appeared to have fewer successful outcomes (43-63% successful outcomes), but there were differences among sites (Fig. 10). Possible explanations for differences among plant species hosts include the degree of concealment provided by the specific plant. For example, the native bunchgrass and plantain are in general, more robust structures, as compared to rhizomatous grasses. The influence of nest host on nest outcome will be assessed in a more quantitative manner in 2017 with Program MARK.

SHLA_JBLM_2016_Final_2March_2017 Page 23

IButtons We deployed iButtons in five Savannah Sparrow nests, two White-crowned Sparrow nests, one Western Meadowlark nest, and four Streaked Horned Lark nests. Half of the lark- surrogate nests, and all of the Streaked Horned Lark nests, successfully fledged at least one young. Two iButtons were ejected by the adults (one Savannah Sparrow and one Streaked Horned Lark) mostly likely because the units were not placed into the nest deeply enough to ensure they were completely hidden. We modified our placement technique and expect to address this problem in the future. We also had two iButtons that malfunctioned and did not collect data and one other that did not record significant temperature signatures. Although sample size was low, our preliminary analyses indicate that iButton temperature data was a good predictor of incubation, foraging bouts, hatch date and nest fate. We also found that iButton presence did not negatively alter parental behavior, risking abandonment of the nest or result in iButton rejection (Potterf 2017). We recommend continuing using iButtons in 2017 to further assess whether this technique is an effective monitoring tool for the lark.

100% 13DIV 90% MAFB 80% GAAF 70% 60% 50% 40%

30% Proportion of Nests of Proportion 20% 10% 0% Abandonment Nest Empty, No sign Nest contents Other of Predation destroyed

Figure 9. Causes of Streaked Horned Larks nest failure at three sites on JBLM, 2016.

Fledglings per nest The mean number of lark fledglings per nests in 2016 was 1.9 (+ 0.1 SE) which was higher than reported in 2015 (1.7 + 0.1), but lower than 2014 (2.5 + 0.2) (Table 5). Productivity (fledglings per nest) was higher at Gray Army Airfield (2.7) than on the native prairie (1.8), or McChord Airfield (1.6). However, it is important to note that sample sizes are small.

SHLA_JBLM_2016_Final_2March_2017 Page 24

Table 5. Breeding summary of Streaked Horned Larks at three sites on JBLM, 2016.

Streaked Horned Lark Breeding Location+ Attribute 13DIV GAAF MAFB Total No. Nests a 22 47 50 119 # Successful Nests (%) 17 (77%) 33 (70%) 29 (58%) 79 (66%) Total No. Eggs 71 170 155 396 Mean Clutch Size (± SE) 3.4 ± 0.2 3.6 ± 0.1 3.4 ± 0.2 3.4 ± 0.1 Hatch Rate (%)b 76 ± 7 90 ± 3 93 ± 1 89 ± 5 No. Nestlings 43 126 108 277 No. Banded 38 97 73 208 Fledglings per Nest (± SE) 1.8 ± 0.3 2.7 ± 0.3 1.6 ± 0.2 1.9 ± 0.1 +13DIV = 13th Division Prairie, GAAF = Gray Army Airfield, MAFB = McChord Airfield; a nests contained at least 1 egg; bproportion of eggs hatched in completed clutches that survived the incubation period.

12 18 16

10 Successful GAAF

14 8 Failed 13DIV 12 10 6 8

4 6 Number Nests Number of Number Nests Number of 4 2 2 0 0 FERO PLLA LULE Grass Other FERO PLLA LULE Grass Other

16 35 14 30

MAFB ALL SITES

12 25 10 20 8 15

6 Number Nests Number of 4 Nests Number of 10 2 5 0 0 FERO PLLA LULE Grass Other FERO PLLA LULE Grass Other Nest Host Nest Host Figure 10: Nest outcome by nest host at 13th Division Prairie (top left), Gray Army Airfield (top right), McChord Airfield (bottom left), and all sites (bottom right), JBLM 2016 (Nest host codes: FERO = Festuca roemeri, PLLA = Plantago lanceolata, LULE = Lupinus lepidus).

SHLA_JBLM_2016_Final_2March_2017 Page 25

4.2.5 Characterizing Human Activities We detected human activities during 33 of 66 (50%) monitoring days at 13th Division Prairie (Fig. 11). Military vehicles (42%), recreational activities (12%), and paratrooper activity (8%) comprised the majority of these events, followed by aircraft activities (e.g., helicopter training, 3%). Relative to 2015, the occurrences of paratrooper events declined, whereas military vehicle activity increased ~38%. We found 26 incidents where vehicles had driven off established roads (17 different survey days), although none of these events resulted in direct impacts to known nests. However, we do not know whether these off-road events resulted in impacts to fledglings. Most of the paratrooper activity occurred in the southwestern portion of the study area, where there were no known nests (Fig. 13). However, on several days during the breeding season, paratroopers were observed landing within close proximity to nesting areas although we documented no direct impacts to known nest locations. However, we do not know whether the paratrooper activities resulted in impacts to fledglings. Staging for the paratrooper activity (e.g., vehicles, buses, etc.) occurred both along the southern portion of the runway and south of the study area. We documented 11 recreation-related disturbances on 8 days, which included the presence of beer bottles, golf balls and shotgun shells located within the priority habitat. Despite the increase in patrolling and signage, recreational users continue to access the site. Based on the presence of the items listed above which were found onsite following weekends, we suspect recreational use increases on weekends. Although the number of recreation incidents increased in 2016, relative to 2015, only one of the 11 incidents involved off-road activity – the majority were observations of vehicles commuting along established roads through the priority habitat. The incidence of recreational use declined after new signage was erected. Determining how these human activities might be impacting larks during the breeding period would require a more systematic assessment. Human activities in occupied lark habitat could result in direct impacts (e.g., individual mortality) or indirect effects (e.g., reduced time incubating, brooding, or feeding; increased predation levels) (Wolf and Anderson 2014).

SHLA_JBLM_2016_Final_2March_2017 Page 26

45%

40% 2013 2014 2015 2016 35% 30% 25% 20% 15% 10%

Percent Percent Survey of Days 5% 0%

Human activities

Figure 11: Human activities at 13th Division Prairie, 2013-2016.

Figure 12: Signage (left), and maps (right) minimized human activities in occupied lark habitat at 13th Division Prairie in 2016.

SHLA_JBLM_2016_Final_2March_2017 Page 27

4.2.6 Noteworthy Species In addition to the lark, two noteworthy species were confirmed breeding within the boundaries of the three nest monitored sites, namely the Common Nighthawk (Chordeiles minor) and Western Bluebird (Sialia mexicana). At least five Common Nighthawk nests were found at McChord Airfield and three nests at 13th Division Prairie, between 20 June and 22 August. An additional notable detection, likely a vagrant or early migrant, was a Short-eared Owl (Asio flammeus) observed on 31 August at McChord Airfield.

5.0 Genetic Rescue at 13th Division Prairie The decline of Streaked Horned Larks in Puget Sound observed during the late 2000s was thought to be driven, in part, by low egg hatchability, which is likely caused by inbreeding (Drovetski et al. 2005, Pearson and Stinson, white paper). This year was the sixth year of a multi-year, multi-partner effort to reverse the decline of the Puget Sound Streaked Horned Lark population by increasing genetic diversity (Wolf 2015). The goal of the project is to increase egg hatch rates by increasing the genetic diversity of the 13th Division Prairie lark population. The objectives of the genetic rescue were to 1) translocate up to 5 clutches for two years to the Puget Sound population (from Oregon), with the hope that 1-2 nests would survive per season, and 7-10 fledglings would survive during the two years; 2) monitor the banded population at 13th Division Prairie and nearby nesting locations to determine survival of larks originating from the translocation; 3) compare pre- and post-translocation hatch rates; 4) conduct a genetic analysis to determine if genes from Oregon enter the South Sound population; and 5) publish experiment and results in a scientific journal. The rescue effort entailed translocating eggs in 2011 and 2013 to 13th Division Prairie from nests at a site in Corvallis, Oregon that was not exhibiting low egg hatchability. Surviving fledglings that successfully breed at the new site should benefit the new population by increasing genetic diversity and improving hatchability. If this experiment is successful in increasing egg hatchability and we can document the mechanism responsible for the success (e.g., inbreeding), this technique could have broader implications to the conservation of birds with small populations throughout the world. In this document, we provide a summary of the 2016 fieldwork relating to the rescue effort. No eggs were translocated to the South Sound in 2016 because of difficulty in matching incubation age and clutch size. To date, a total of 20 eggs have been translocated to 13th Division Prairie since the genetic rescue effort was initiated - twelve eggs (4 clutches) in 2011 and 8 eggs (three clutches) in 2013. No banded birds that originated from Corvallis translocated clutches were observed at lark nesting sites outside of 13th Division Prairie (i.e., McChord Airfield, Gray Army Airfield, JBLM Range 76 of Artillery Impact Area, Olympia Airport, or Sanderson Airfield) (Fig. 1). The Oregon translocated nestling (i.e., the Oregon Male) that returned to 13th Division Prairie as an adult male in 2012 and bred successfully in 2013, 2014 and 2015, returned again in 2016, and had at least two nest attempts that successfully fledged a total of 6 young. To date, the Oregon Male has had at least 13 nest attempts, and we documented 100% hatch rates for all 20 eggs in the nests that reached the nestling stage (n = 7). At least seven of his 13 nest attempts were successful, resulting in at least 18 fledglings. Based on annual juvenile return

SHLA_JBLM_2016_Final_2March_2017 Page 28

rates (0.25), we would have expected at least four young to return. However, only one juvenile of 18 (6%) that survived from the translocated Oregon male’s third nest attempt in 2014 was resighted as an adult male at 13th Division Prairie in early May 2015. It is unknown whether the failure of young to return and breed successfully could be attributed to the Oregon genes. For example, these birds might be predisposed to be resident and therefore not survive because they remain in the Puget Sound area instead of migrating, and thus, could be maladapted to South Sound weather. An alternate explanation is that the young birds survive and migrate but do not return to the south Sound after the winter. A genetic analysis was initiated on genetic material collected from 2011 through 2014 (195 feathers, 60 eggs, 13 nestlings, and 2 adults). Due to low quality and quantity of DNA extracted, an additional round of sequencing was performed, and preliminary results indicate structuring among sites. In addition to 2015 samples (n = 186), we collected/salvaged 228 samples in 2016 (central retrix) from 185 individuals, 25 eggs, and 18 nestlings). To date, we have collected 506 feathers, 131 eggs, 41 nestlings, 6 fledglings, and 5 adults for the genetic analysis, and results will be forthcoming. The genetic analysis for this study is being led by WDFW. It is ongoing and aims to answer the following questions:  Are there genetic differences between Oregon and Washington Streaked Horned Larks?  Is there an association between genetic differences and hatchability? Are the genes from hatched WA and OR eggs/clutches more similar (somewhere in the genome) than WA unhatched eggs/clutches and OR hatched eggs/clutches?  Is there evidence of inbreeding in the WA and OR populations?  Is there an increase in genetic diversity in the WA population receiving the OR genes?

The genetic analysis will assess the potential mechanism for any observed change or lack of change in egg hatchability, by comparing genetic diversity (levels of heterozygosity, inbreeding coefficient) before and after the egg swapping experiment. In summary, 2016 was the sixth year of the Genetic Rescue Project. We successfully translocated Oregon clutches to the Puget Sound population in 2011 and 2013, and confirmed that a male originating from an Oregon translocated clutch in 2011 survived its fifth winter and returned to breed at 13th Division Prairie. We confirmed that the translocated Oregon male successfully produced at least 18 fledglings and based on our juvenile survival rates over this period (0.25), we expected at least 4 fledglings to survive their first winter and reproduce the following year. However, we have only detected one juvenile that survived the winter, and this bird was only observed briefly in early May 2015. Nonetheless, if the Oregon Male’s offspring survives and breeds successfully at 13th Division Prairie in future years, the local population of Streaked Horned Lark could be rescued by this single bird, which can lead to improved fitness and reduced extinction risk (Waite et. al. 2005). To account for temporal and seasonal aberrations that affect lark productivity and survivorship, several years of translocating a greater number of egg clutches per year, or considering alternative approaches, might be necessary to successfully introduce genetic material into the population. However, the results of the genetic analysis, and higher hatch rates for the 13th Division Prairie population, should be considered prior to any future actions.

SHLA_JBLM_2016_Final_2March_2017 Page 29

6.0 Juvenile and Adult Return Rates Juvenile and adult survivorship has the largest influence on Streaked Horned Lark population growth rate (Camfield et al. 2011). While there is published and important information on annual survivorship rates for larks (Camfield et al. 2011, Pearson and Hopey 2008, Wolf et al. 2016), those models do not incorporate time dependence and we still lack information on what time period is most limiting to lark populations (e.g., immediately post- fledging, when parents no longer tend juveniles; over-wintering). We also lack information on the sources of mortality (e.g., predation, airstrike, mowing). Answering these questions will be critical to developing conservation strategies aimed at recovering lark populations. To begin addressing this information gap, our objectives were to collect resight data of color-banded individuals to: 1) estimate annual return rates of juveniles and adults, 2) identify causes of mortality, and 3) assess natal dispersal rates. Our long-term goal is to use this information to develop management recommendations aimed at increasing Streaked Horned Lark juvenile and adult survivorship in south Puget Sound. As part of a complementary project funded by USFWS and the Army Compatible Use Buffer Program (Slater 2016), we also assisted with a radio-telemetry project to identify causes of juvenile mortality, determine factors associated with survival, and evaluate habitat use. This is a 3-year study that will continue through 2018 breeding season. Results from the first year of the project are available in Slater (2016). 6.1 Methods We have been banding lark nestlings with unique color leg bands (see Section 4.1) and conducting resight surveys at JBLM since 2011. Fieldwork has been conducted during the breeding season at three sites: 13th Division Prairie (2011 to 2016) and Gray Army and McChord Airfields (2013 to 2016). In 2015, we affixed blue automobile tape around USFWS bands to increase the number of unique color-band combinations available. We applied the blue tape around the USFWS bands of 88 of 143 lark nestlings in 2015. We found that the tape remained attached during the immediate post-fledging period in 2015 (62 of 88 individuals were resighted as fledglings, Wolf et al., 2016), but the tape did not remain on the bands over the winter. As a result many of these birds that survived the winter were unidentifiable from other birds with the same color band combinations. We confirmed that the tape fell off through: 1) field testing the blue tape in 2016; 2) verifying the USFWS band numbers on larks that were found dead at McChord Airfield (see section 6.2); and 3) confirming USFWS band numbers on adult birds with spotting scopes. As a result, we discontinued the use of blue tape in 2016. For estimating juvenile return rates in 2016, we only used 62 individuals whose band combinations were unique.

6.1.1 Resight Data We conducted serpentine surveys, modified line transects, designed to resight (recapture) marked juveniles and adults. They also helped assess spatial distribution and guide nest monitoring activities. These surveys were conducted four times during the year (between 18 April and 4 August) and complemented abundance surveys (section 2.0). The placement, orientation and spacing of the serpentine transects were designed to survey all suitable habitat within the study areas at three sites (13th Division, McChord Airfield, and Gray Army Airfield).

SHLA_JBLM_2016_Final_2March_2017 Page 30

The distance between transects in the serpentine surveys were spaced 100 m apart. Total transect length was 17.1 km, 11.5 km and 11.0 km at 13th Division Prairie, Gray Army Airfield, and McChord Airfield, respectively. Observers mapped and tabulated all larks seen or heard, identified animals by sex and age (when possible), recorded band combinations (if present), and documented secondary behaviors and environmental variables. We also collected resight data opportunistically during nest monitoring activities.

6.1.2 Return Rates We calculated annual return rates for adult and juvenile larks in the JBLM population by calculating the number of individuals alive in breeding season t +1 divided by the number of individuals alive, or nestlings that successfully fledged (for juveniles), in breeding season t. We defined the breeding season resight period from April – June to minimize bias, but long enough to ensure a high likelihood of resight given the animal was alive. For juveniles, we calculated an overall return rate and also compared return rates by natal site and sex for juveniles returning as adults. 6.2 Results and Discussion

6.2.1 Resight Data In 2016, serpentine surveys were conducted on 12 days, representing over 43 hours of survey time. We recorded 386 lark detections at the three sites and many of these were repeat detections of the same bird at a given site. Of the 386 detections, 85 (22%) were color-banded birds: 69 were adults (53 individuals), and 16 were juveniles (14 individuals). We also captured 106 band resight events during nest monitoring activities (section 4.0), and another 6 events during the banded bird resight surveys (section 5.0).

6.2.2 Return Rates In 2016, we detected 68 color-banded adults in our study areas, 45 (66%) males and 23 (34%) females (Appendix 2). This indicates that 37-39% of the entire estimated JBLM lark population (87-92 pairs) was color-banded in 2016. Of the 68 adults resighted, 43 (63.2%) were first year breeding birds (25 male and 18 female); this is probably due to the large number of young banded in 2015. At 13th Division Prairie, one adult male returned to his natal breeding site for his seventh breeding season, and two adults (one male and one female) returned for their fifth breeding season. As described earlier, we affixed blue automobile tape around USFWS bands to increase the number of unique color-band combinations available. As a result of the blue tape falling off, there were at least 10 adults with either duplicate band combinations, or where we were unsure these animals had returned as after-second year breeding birds. Of the 35 breeding adults detected in 2015, at least 22 were resighted in 2016, yielding a minimum annual return rate of 0.63, which was similar to the return rate in 2015 (Table 6). Again, this value for 2016 is likely an underestimate, because we were uncertain about the ages of at least 10 banded adults. The overall annual adult return rates for adults since 2010 is 0.68. These rates include four male larks which were found dead during the early breeding season on McChord Airfield, presumably struck by aircraft (see below).

SHLA_JBLM_2016_Final_2March_2017 Page 31

Overall juvenile annual return rate in 2016 was 0.39 with 24 of 62 nestlings banded in 2015 resighted in 2016 (Table 7). This was the highest return rate recorded since monitoring was initiated in 2011. Although sample sizes were small in 2016, juvenile return rate was highest for young at Gray Army Airfield (0.53), followed by McChord Airfield (0.36), and 13th Division (0.14) (Table 8). A total of 260 uniquely color-banded larks have successfully fledged on JBLM between 2011 and 2015 (Table 7), and the overall return rate for juveniles is 0.25 (65 of 260 returned), which is in the mid-range reported for other granivorous landbirds (range = 8- 42%, Verner et al. 1998). More males were resighted compared to females (43 versus 22), presumably because they are more philopatric to their natal site. In general, we have little information about the major causes of mortality for adult and juvenile larks. However, in 2016 four adult males, all first-year birds, were found dead on McChord Airfield, presumably struck by aircraft (Christopher Lang USDA, pers. comm.). Three of the four larks were found within 400 m of the intersection of taxiway Charlie and runway (5 May, 13 May, and 27 June); the fourth lark was found on the runway centerline, south of taxiway Delta. First year breeding males may be more vulnerable to airstrikes because they are trying to acquire and maintain territories. These inexperienced males may conduct flight displays, which may occur >100 m above the ground, more frequently than established males making them more vulnerable to aircraft strikes. They are also likely less experienced near runways. Finally, this spike in mortality events may be due to the increased number of bird at the site (20-25 pairs this year vs. 15-18 last year) or the aggregation of six breeding territories in the vicinity of taxiway Charlie (Figs. 3 and 6). In August and September 2015, we observed five banded larks with tumor-like lesions that were identified as avian pox (Kristen Mansfield, Washington Department of Fish and Wildlife, pers. comm.). Out of the 5 banded larks that presented with visible lesions, two were confirmed alive and well in 2016, returning as breeding females at McChord Airfield. One of these larks that returned was a female who produced 2 successful fledglings. Two other color- banded larks may have also returned, lesion-free, but we are uncertain because they share band combinations with larks that had blue tape applied over their silver bands. As reported previously, this blue tape has fallen off, and this has left these birds essentially double banded. These results suggest that larks are capable of acquiring pox and surviving. Also interesting to note, we documented two color-banded larks with severe limps in fall of 2015, but without visible lesions. One of these larks was confirmed alive at Gray Army Airfield in 2016. Most returning first-year larks in 2016 returned to their natal site, although 10 of 43 (23.2%) dispersed to non-natal sites (Appendix 2). Eight dispersed from Gray Army Airfield to other sites (seven moved to McChord Airfield, and one dispersed to Olympia Airport). Two dispersed from McChord Airfield to other sites (one moved to Gray Army Airfield, and one moved to 13th Division Prairie). Curiously, we found no first-year adults that had dispersed from 13th Division Prairie, perhaps due to the extremely low juvenile return rates from that site. We detected two larks at Olympia Airport in 2016 – both originated from JBLM, and one was a first year first female that had dispersed from Gray Army Airfield (Appendix 2). We detected four color-banded larks at Range 76 (3 of 4 were returning adults), two that originated from Gray Army Airfield, one that dispersed from 13th Division Prairie; the third was an incomplete band combination. We did not detect any banded larks at Range 50.

SHLA_JBLM_2016_Final_2March_2017 Page 32

Table 6. Number of adult larks resighted and return rates, JBLM 2011 - 2016.

# adults # adults Years Return Rate (%) returned previous year 2010 - 2011 5 Unknown Unknown 2011 - 2012 3 5 60.0 2012 - 2013 5 6 83.3 2013 - 2014 5 7 71.4 2014 - 2015 10 16 62.5 2015 - 2016 >22* 35 62.9 Average (2011 – 2016) 68.0 * Asterix indicates a minimum number of adults (see text)

Table 7. Number of color-banded lark nestlings that successfully fledged and their return rate after their first winter, JBLM 2010 - 2016.

# nestlings # adults surviving their Years Return Rate (%) banded* 1st winter 2011 15 3 20.0 2012 13 2 15.4 2013 66 11 16.7 2014 104 25 24.0 2015 62 24 38.7 2016 208 TBD TBD Total 485 Average = 23.0 *excludes 6 nestlings that died in the nest, 81 nestlings with duplicate bands (see text)

Table 8. Number of color-banded lark nestlings that successfully fledged and their return rate by sex after their first winter at three sites, JBLM 2015-2016.

# nestlings banded # adults surviving their 1st winter Total, Return Site* in 2015 Male Female Rate (%) 13DIV 7 1 - 14.3 MAFB 36 7 6 36.1 GAAF 19 4 6 52.6 Total 62 12 12 38.7 *Site Codes: 13DIV = 13th Division Prairie, GAAF = Gray Army Airfield, MAFB = McChord Airfield

7.0 Optimizing Site Co-Use The goal of optimizing site co-use is to simultaneously implement conservation measures to enhance the survival and recovery of the Streaked Horned Lark, while also

SHLA_JBLM_2016_Final_2March_2017 Page 33

facilitating JBLM’s mandate for military training. Our objectives are to use the information collected during our transect surveys, nest monitoring, and survivorship research to support decision-making for management objectives at occupied sites while also guiding minimization of direct impact to larks. We worked directly with lark-occupied site managers and JBLM Fish and Wildlife to communicate lark nest status at a minimum of a weekly basis. We also conducted a flush distance analysis aimed at understanding the spatial considerations when trying to minimize human-caused disturbances to young of the year. 7.1 Breeding Phenology and Fledgling Vulnerability Below, we summarize the timing, duration, and spatial arrangement of Streaked Horned Lark life stages in south Puget Sound (Table 9). We also provide our general evaluation of the vulnerability level of each life stage. In 2016, we documented the earliest clutch initiation date for Streaked Horned Larks in the South Puget Sound – 28 March. This date was estimated after the detection of a juvenile lark on 21 April. We assume that the initiation of early laying was associated with a spell of early-spring warm weather, and that only a few pairs might have initiated early clutches. Most 2016 nests were initiated in three periods: mid-to-late-May, mid- June and mid-July. However, this early initiation illustrates the variability in breeding timing, which is influenced by many factors, including seasonal/climatic conditions. In mid-March 2016, Western Washington experienced above-average temperatures, which likely prompted the birds to nest early. These early nesting events might become more frequent as the predicted phenological responses of birds to global climate change are advanced arrival dates and earlier laying (e.g., Fletcher et al. 2013, Cotton 2003). Monitoring may need to start earlier in the year if this pattern continues.

Table 9. South Puget Sound Streaked Horned Lark life stages and vulnerability

Significant Stage Spatial Ground Life Stage Date (2016) Duration arrangement Vulnerability First detection Throughout Adult Arrival ~ 2 months Low 4 Feb sites First egg laid Nesting (Eggs, ~28 March late-March Confined to Moderate - Nestlings) First hatch to mid-May nest location high date ~11 April 0-2807 m First fledge early April Fledgling from natal High date ~18 April to late Aug nest Last date of Throughout Independence independence After 1 Sept Low sites ~ 1 Sept

SHLA_JBLM_2016_Final_2March_2017 Page 34

7.2 Communications with JBLM Airfield and Training Area Managers We communicated results of nest monitoring activities to airfield and training area managers as often as necessary to provide managers with up-to-date information (often daily, but on a weekly basis at a minimum). We provided airfield and training area managers with spatially explicit maps that illustrated nest site locations, recommended stay out areas (BANAs, or Buffered Active Nest Areas, with buffer radii of 118 m), and vulnerability periods for the three nest monitoring sites. On each map, nest locations were identified as either 1) active (e.g., located), or 2) inactive (fledged or failed). Active nests were also assigned a vulnerable period date, defined as 14 days post fledging, a period when fledglings were flightless or considered weak flyers and thus unable to avoid military and management vehicles (e.g., mowers). These fledglings were also likely dependent on adults for food provisioning. For the vast majority of the breeding season, the BANAs were observed and avoided by airfield mowing and maintenance crews. However, our biologists recorded a few instances at McChord Airfield when mowers were observed within the BANAs, and maintenance vehicles had driven off the runway into vulnerable areas. When notified, McChord Airfield BaseOps were responsive and immediately informed the operators to cease their activities within the BANAs. We recommend that the mower and maintenance personnel exercise more vigilance when working within close proximity to the nest areas (see Section 9.1). CNLM also worked closely with Gray Army Airfield personnel and JBLM Fish and Wildlife to minimize potential impacts to active nests during a 2-night paratrooper training. CNLM deployed 8 chemical lights around each active nest (radius of 10-15 m) in the early evening prior to the training activities. The chemical lights remained lit through the evening, with the goal of being readily detectable by paratroopers so they could avoid these nest locations. CNLM biologists deployed the lights in the early evenings of 15 and 17 June, around 10 active nests at the egg/nestling/or recently fledged stage. We retrieved the lights on the mornings after the evening trainings, and found that the status of all nests, save one, remained unchanged. The one event was the discovery of one dead, 7 day-old fledgling within close proximity to one active nest; four other young in the same nest had apparently fledged. Biologists noticed evidence of a vehicle track very close to the nest. However, we cannot be sure that the training activities resulted in the direct mortality to this individual. We believe the realignment of the drop zone (and staging areas) at 13th Division Prairie continue to minimize the level of disturbance to breeding pairs (Fig. 13). In 2016, as in 2015, the majority of the paratrooper activities were concentrated in the western and southwestern portion of Training Area 14, where no known nests were located. In addition, the three staging areas used for the drops were located south of the drop exclusion zone and on the southern portion of the runway. As a result, paratroopers returning to the staging areas tended not to walk through the majority of the known nesting areas. However, off-road military vehicle activity (often associated with the paratroop drops) was recorded on ~25% of the survey days and should be curtailed during the breeding season to minimize potential impacts. Although recreational users continue to access the 13th Division Prairie, the majority of the incidents observed were vehicular traffic commuting along established roads which pose a risk for flightless juveniles. We believe that the presence of the signage, and response of Down Range Police forces to trespassing activities continue to discourage and reduce the incidence of recreational users, relative to 2013 (Fig. 11).

SHLA_JBLM_2016_Final_2March_2017 Page 35

Overall, we believe that the increases in reproductive success in 2016, 2015 and 2014, relative to 2013, may be partially explained by the reduction in direct impacts to lark nests as a result of: 1) avoidance by mowing activities at the airfields; 2) altered military training (e.g., paratroop training at 13th Division Prairie); 3) deployment of chemical lights for evening training activities at Gray Army Airfield; and 4) increased signage at 13th Division Prairie to minimize recreational use. For the past two years, we have not documented any nest failures at either the egg or nestling stages that resulted directly from mowing impacts. We believe that this was due, in part, to the diligence of the airfield managers conveying, in a timely manner, the spatially explicit maps to the mowing equipment operators.

Figure 13. Observed paratrooper drop zone and staging areas, 13th Division Prairie, 2016.

SHLA_JBLM_2016_Final_2March_2017 Page 36

7.2 Fledgling Flush Distance In 2015, we initiated work to better understand how young birds respond to disturbances and management activities that could result in trampling (e.g., mowing, vehicle and human traffic). Data were collected during the nest monitoring activities and serpentine transects (Sections 4.1.1, 6.1.1). We estimated flush distance (i.e., distance from observer when juvenile flushed) and the distance flown by color-banded juvenile larks. When observers found juvenile birds, and were able to confirm band combinations, the bird was approached on foot at a brisk pace. In general, observers confirmed band combinations from a distance and then initiated the flush experiment. However, on occasion, some individuals were found opportunistically by flushing and then reflushed, once their band combinations were confirmed. We do not know whether a second flushing event may have caused individuals to flush more quickly or flush less quickly than if they had not just previously been flushed. However, we believe that these data are biologically relevant because these animals might be subject to subsequent disturbance events in occupied sites when, for example, mowing vehicles make repeated passes at the airfields. The distance at which the young bird flushed from the observer, and the distance flown was recorded. If the bird remained in place upon approach, flush distance was recorded as 0. Only one flush distance was recorded for each individual for a given survey day. However, these data include flush distances for the same individual on multiple survey days. We decided that for flight-capable juvenile larks to safely avoid tramping, they were required to have the following responses to on-foot disturbances: 1) the disturbance must elicit a flight response (i.e., stress response must not be passive); 2) flush distance must be >2 m, allowing sufficient time for the juvenile to engage in flight; and 3) distance flown must exceed 5 m. These decisions were based on the following: 1) adequate time and distance for an active defense response (i.e., physiological flight response), 2) operating speed of a tractor pulling a deck mower (5 mph), and 3) flight distance necessary to avoid a 2-3 m wide tractor mow deck, while also allowing enough time for individual to re-flush. In 2016, we collected an additional 35 flush distance data points for color-banded juveniles who ranged in age from 4 to 82 days old post-fledgling. We pooled 2015 and 2016 flush data (n = 144), and calculated the mean flush distance of all encounters to be 11.5 m (± 0.7 SE). Flush distance was positively correlated with fledgling age (Pearson Coefficent of Correlation = 0.47, t-Stat = 6.39, df = 142, P < 0.001) (Fig. 14). The majority (117 of 144, 81%) of flush data were juveniles > 10 days post-fledging. This can be explained by the difficulty of detecting younger juveniles, which are extremely camouflaged, remain motionless, and are thus more difficult to locate. We found that on-foot disturbances resulted in a flight response for the majority of the events. In response to observers approaching, most juveniles took flight (123 of 144, 85.4%), which was less than responses recorded by Pearson and Hopey (2004) for adult larks (95%). Of the 144 juvenile disturbance events, 21 (14.6%) were individuals that remained in place and did not flush when approached. Of the 21 that did not flush, seven were younger than 4 days old and were probably unable to fly. The remaining 14 larks that did not flush ranged in age from 4 to 27 days old post-fledgling and were likely able to fly. Thus, even though fledglings might be

SHLA_JBLM_2016_Final_2March_2017 Page 37

flight capable, there appears to be a strong selection for relying on camouflage and remaining motionless to evade predators. Our preliminary results indicate that juveniles ≤ 10 days post-fledging do not flush sufficiently to avoid trampling (Fig. 14). Only 4 of 27 juveniles ≤ 10 days post-fledging were able to flush at distances ≥ 2 m. After 10 days, juveniles flushed at a mean of 13.8 m (± 0.7 SE, n = 117), which we considered was a sufficient distance to avoid trampling from foot traffic, and vehicular traffic. Yet, there were 13 juveniles (11%) > 10 days post-fledging that either stayed in place or flushed at distances < 5 m. This suggests that our conservation action of retaining buffers around nest sites for the two-week period post-fledging might be a sufficient period to protect the majority of (but not all) vulnerable individuals that remain within the 118 m nest buffer (BANA). However, we know that juveniles < 10 days post-fledging do move beyond these buffers, and it is these animals that would be vulnerable to trampling. For example, in 2015, we found that of 48 color-banded juveniles that were ≤ 10 days post-fledging, 14 (29%) were found further than 118 m from their natal nests (Wolf et al., 2016). Consequently, the nest buffers would not protect these animals from trampling. Our data also indicates that young birds do not flush as readily to on-foot disturbances as adult larks (11.5 m for juveniles versus ~17 m for adults, Pearson and Hopey 2004). Our data suggests that juveniles only begin to respond similarly to adults at approximately three-weeks post-fledging. We collected distance flown during 27 events, and calculated the mean to be 71.2 m (± 4.0 SE). These data only included juveniles that flushed, and did not include those that remained in place, and did not flush. We found no relationship between fledgling age and distance flown, likely because of the small sample size (Pearson Coefficent of Correlation = 0.27, t-Stat = 1.4, df = 26, P = 0.08). Of the 27 events, 25 (93%) were individuals >10 days old, post fledging. All distances flown exceeded 5 m, and we recorded only one event where distance flown was < 10 m, which entailed a ten day old juvenile. We recorded only one event which involved a juvenile <10 days old, and this 8 day old, post-fledgling animal was able to fly a distance of 45 m. Our data indicates that average distance flown by juveniles is greater than distances flown from disturbances by adult larks (40m for adults, n = 55, Pearson and Hopey, 2004). This might be explained by the propensity for breeding adult larks to remain committed to their breeding territories, and nesting locations. Our data suggests that flight-capable juvenile larks are proficient at flying distances sufficient enough to escape from the path of a mower, or military vehicle (> 5 m). Yet, these data also illustrate that disturbances do not always elicit an active defense response in younger larks, particularly those younger than 10 days, post-fledging. It appears that there is a strong selection for young larks to rely on crypsis for evading predation, which makes them extremely vulnerable to trampling. These data support the conservation action of protecting buffers around nests (recommended stay-out areas) for 2 weeks after the nest fledges successfully. As mentioned above, we do not know if a second flushing may affect flush response, and whether a second flush event may cause individuals to fly shorter or greater distances than if they had not just previously been flushed. This has relevance when considering how juveniles might respond to repeated disturbances. For example, tractors mowing airfield vegetation perform repeated sweeps to widen the mowing swaths until all vegetation is cut to a suitable height. Thus, it is entirely feasible that the same juveniles might be flushed more than once during the course of the management activity, and these birds would have to flush repeatedly,

SHLA_JBLM_2016_Final_2March_2017 Page 38

in a timely manner, and fly an appropriate distance to avoid trampling. We also do not understand among-site juvenile lark response at sites that receive different levels of human use. Because birds nesting close to humans tend to tolerate more disturbances than birds nesting in remote areas (e.g., Gabrielsen et al. 1985), we assume that larks at the airfields will flush at closer distances than larks at native prairies. Further research would be necessary to understand juvenile response to multiple disturbances, and whether there are differences among sites.

45 40 35 30 y = 0.1984x + 5.79 25 R² = 0.2233 20 15

Flush Flush distance (m) 10 5 0 0 10 20 30 40 50 60 70 80 90 Fledgling Age, # days post-fledging

Figure 14: Distance, and age, of lark fledglings to observers when flushed, 2015-2016.

8.0 Habitat Management at 13th Division Prairie We continued habitat enhancement activities for Streaked Horned Lark on priority habitat in Training Area 14 in summer/fall/winter 2015 and 2016. Comprehensive details of the strategy and implemented actions were included in a separate report (Kronland et al. 2016). In summary, the following actions were performed:  Mowed Scotchbroom (Cytisus scoparius) and blackberry (Rubus sp.) on 55 acres  Chemically treated blackberry on ~3.2 acres  Manually and chemically treated Queen Anne’s lace (Daucus carota) on 8.1 acres  Chemically treated Oxe-eyed Daisy (Leucanthemum vulgare) and Scotchbroom on 13.0 acres  Chemically treated Rough Cats-ear (Hypochaeris radicata) on 10.3 acres  Chemically treated Trifolium subterranean on 1.4 acres  Chemical treatment of non-native grasses (e.g., Agrostis capillaris) on >11.3 acres  Removal of Douglas fir (Pseudotsuga menziesii) on 0.04 acres  Seeded ~75.2 lbs. of native bunchgrass (Festuca roemerii) and ~16.2 lbs. of native forbs on approximately 30.2 acres.  Conducted prescribed fire on 372.3 acres in 2015, and 308 acres in 2016

SHLA_JBLM_2016_Final_2March_2017 Page 39

9.0 Population Management Recommendations The goal of this section is to outline management recommendations that would provide for the increase in, and recovery of, the JBLM Streaked Horned Lark population. The objectives include 1) identifying ongoing actions that continue to have conservation value, 2) recognizing the greatest information gaps hindering recovery, and 3) proposing specific actions addressing these information gaps. 9.1 Coordinate Management Actions Closely with Nest Status We believe that the period prior to the onset of early nesting presents the best opportunity to engage in maintenance activities (e.g., mowing) at occupied sites. This pre- breeding period, generally mid- to late-March, is before the majority of early nests are in the incubation and early nesting stage. At this time, there are no fledglings on the ground. However, the actual timing of this period varies depending on weather conditions, and when nesting activities are initiated. The early initiation of a few clutches in 2016 suggests that maintenance activities may need to be extended earlier in years with warm spring temperatures, to protect early nesting pairs (early-March through early-April, Fig. 8). In 2016, there was a slight lull in nest initiations in late-May/early-June and late-June. However, during this time, vulnerable fledglings were still dependent on adults, and new nests were reinitiated. In addition, when fledglings move beyond the recommended stay-out areas, we have little confidence in knowing where the young of the year are distributed and the maps showing nest locations are not effective at the post-fledging stage. Thus, there was no period during the mid-late breeding season where nests and fledglings are not vulnerable. Following, the vulnerability period extends from the time when nests are initiated (early-April) until young of the year are strong flyers foraging independently (early September). Our recommendation is to continue to use a combination of survey techniques to avoid impacting Streaked Horned Larks. Early in the breeding season (March/April), we suggest conducting serpentine surveys, and use these data to map territories/use areas. If land managers are interested in protecting all adults, nests, and vulnerable fledglings, we recommend that human activities be reduced or avoided within the defined territories and use areas. Buffering territories/use areas could protect nest sites, as well as sufficient area such that human activities will not adversely affect adult nest-tending behavior. Coupling nest- monitoring activities with identifying use areas could further ensure that the majority of nest sites are avoided. Of critical importance is the conveyance of these survey and monitoring results to site managers, and appropriate field personnel. We recommend that site managers continue to be provided regular updates of nest status and locations, in the form of spatially explicit maps of each occupied site that indicate nest locations, appropriate buffers, and vulnerability periods. Inputting nest coordinates and recommended stay-out polygons into handheld global position system (GPS) units could also be used to advise mower operators of which areas to avoid. In summary, a combination of monitoring techniques (nest monitoring and transects) and excellent communication with site managers are the best management practices to minimize human-caused disturbances.

SHLA_JBLM_2016_Final_2March_2017 Page 40

9.2 Continue Lark Monitoring at Occupied Sites Locating and monitoring Streaked Horned Lark nests remains the most effective method to direct management at occupied sites aimed at precluding negative impact to breeding larks. We believe that nest locating activities identified the vast majority of nests at all three sites in 2016, although we know a few nests went undetected. We continue to believe that adult locations from abundance transects are not a suitable proxy for determining nest locations. A preliminary home range analysis using 2016 abundance transect locations showed that the data overinflated general use areas at 13th Division Prairie (+85%) and Gray Army Airfield (+17%) and slightly underestimated use at McChord Airfield (- 9%), relative to the Territory Mapping data. Furthermore, the abundance transect generated home ranges also failed to capture 5 nest locations, whereas the general use areas from the territory mapping effort captured all known nests. However, although our general use area mapping efforts captured most nest locations, nests were missed, as evidence by unbanded young of the year. The territory mapping effort might also be more time consuming than locating and monitoring nests. Furthermore, using the larger site-level home range estimates (general use areas), which combine all territories, as “recommended stay out areas” would give less flexibility for management activities (e.g., mowing), particularly at Gray Army Airfield. This is because the vast majority of the airfield would be mapped as a home range.

9.2.1 Territory mapping We recommend continuing the home range analysis and believe that these data could have positive conservation implications, including directing human activities, and guiding habitat enhancement and restoration activities. We have identified several key questions that could be answered by the effort:  Home range consistency – are the home ranges consistent amongst years?  What degree of home range overlap is possible – is there a threshold for the number of lark use areas within a defined space? And would approaching this threshold have an effect on reproductive fitness?  At what time scale might “new” habitat be needed to accommodate a predicted increase in the population? (e.g., when would carrying capacity be reached at the airfields, which are showing the greatest increases in the number of breeding pairs)  Are there optimal foraging locations, and could these areas guide where training should be confined to minimize impacts.  Do larks modify their home range patterns over the breeding season, perhaps in response to changes in foraging resources as vegetation senesces.

9.2.2 IButtons We recommend continuing to assess whether iButtons (temperature data loggers) are an effective lark monitoring tool. Results from the pilot study indicate that larks accept iButtons in nests and that iButtons accurately assess survival and failure. Continued research should focus on whether iButton data yields more accurate estimate of nest survival, compared to our existing protocol which relies on nest checks every 3-5 days. IButtons could also be used to better evaluate the effect of military training on lark nesting. For example, iButtons could measure periods of both incubation and foraging bouts, which could influence egg or nest

SHLA_JBLM_2016_Final_2March_2017 Page 41

survival, in relation to military training activities. Furthermore, iButtons could also reduce the intensive nest monitoring effort, and produce cost savings.

9.2.3 Occupancy surveys in new occupied sites We also recommend conducting occupancy surveys in the “new occupied sites” early in the breeding season to determine if these sites continue to be occupied in 2017. These sites are Range 52 and the western portion of Range 50, in the Artillery Impact Area. If these sites are determined to be occupied, abundance transects should be established in these areas.

9.2.4 Target trap adults We recommend target trapping certain adult larks in 2017 to resolve issues with the blue tape (over the USFWS) falling off (section 6.2). Of the 68 banded adults detected in 2016, we identified 34 individuals with problematic band combinations. Based on an overall adult return rate of 68%, we anticipate that we would only have to capture 23 animals. We suggest employing mist nets, bow-nets, or potter traps around nest sites, or use a combination of mist- nets and taped-playback to capture adults. Development of a successful adult trapping technique would also facilitate our survivorship research because we could focus banding efforts on adults rather than banding every nestling, which is an extremely time-intensive effort and juvenile survivorship rates are low. 9.3 Identify Nest Predators Evaluating the need to control predators, and the effects of predator management on population trends, was ranked in the top 10 actions (#8) in the Streaked Horned Lark action plan. Because predation is the primary reason for nest failure on JBLM, we recommend designing and implementing a study to identify nest predators at the three nest monitored sites; and then evaluate the need for, and/or design, a predator management and monitoring plan that can be applied in a systematic, scientific manner. We recommend two techniques to address this question: 1) placement of remote wildlife cameras at nest sites; and 2) attaching telemetry transmitters to nestlings prior to fledging. These data will help identify causes of nest failure, and mortality during the post- fledging period. This information will help guide management decisions when determining whether predator removal/control could reduce mortality. We believe that this effort would have great conservation value for the lark both on JBLM, and throughout its range. 9.4 Convert Vegetation at Airfields Although we documented that all known nests for the past two years were unaffected by mowing activities at the airfields, we are less certain about whether mowing resulted in direct mortality of fledglings. Furthermore, because there is no apparent period in the breeding season where nests and fledglings are less vulnerable, we do not recommend unrestricted mowing during the breeding season. Rather, we suggest pursuing three habitat management strategies to minimize mowing-related impacts, which should increase lark productivity. The first approach is a short-term strategy to lure larks away from high conflict areas at McChord and Gray Army Airfields. Research indicates that nest distance to runways may negatively influencing daily nest survival rate (Peters and Allen 2010). In addition, adult larks have been struck by aircraft (see Section 6.2.2). Thus, we suggest creating attractive lark habitat

SHLA_JBLM_2016_Final_2March_2017 Page 42

in areas within the airfield that receive less training use, far removed from high human-conflict areas of the airfields (e.g., runway aprons). For example, grassland habitat adjacent foxtrot row and in the southern portion of the overrun at McChord Airfield could be treated with multiple herbicide applications, which may reduce the need for mowing during the breeding season. Enhancing habitat suitability for larks in these lower-conflict areas could make them more attractive to breeding and foraging larks and may draw them from higher conflict areas. The second approach is a long-term strategy to replace airfield vegetation with native bunchgrasses. Replacing the tall sod-forming grasses with native perennial bunchgrasses (e.g., Festuca roemerii) would reduce the need for frequent mowing to stay within prescribed height restrictions. Doing so, should reduce fledgling and nest mortality. This could be accomplished with multiple habitat enhancement techniques such as prescribed fire, followed by herbicide applications, and broadcast seeding. A third, long-term, approach is to lure larks from the active airfields by restoring and enhancing habitat adjacent to the active airfields. Suitable lark habitat within close-proximity to the current occupied sites is more likely to lure larks than would be habitat created further away. The ultimate goal would be to draw larks from the airfields to these areas (coupled with dissuasion techniques), which would reduce lark numbers on the airfields, and thus reduce the risk of aircraft-lark strikes. Habitat enhancement strategies have already been developed for, and implemented at, Close-In Training Area F (CTAF) located immediately south of Gray Army Airfield. The outcome of CTAF habitat activities could provide guidance for conducting similar work in the South Approach Zone at McChord Airfield (SAZ). Replacing the tall exotic plants (e.g., Scotchbroom) in the SAZ with native perennial bunchgrasses would also minimize aircraft- strike risk and reduce the burden of wildlife management activities in these densely vegetated areas adjacent the airfield (Chris Lang, USDA, pers. comm.). 9.5 Evaluate Effect of Disturbances Human-caused disturbances were detected during 33 of 66 (50%) monitoring days at 13th Division Prairie (see Section 4.2.5). Military vehicles comprised the majority of these disturbances. Relative to 2015, military vehicle activity increased ~38%. We found 26 incidents where vehicles had driven off established roads through the prairie, although none of these events resulted in direct impacts to known nests. Impacts of these, and other, disturbances on lark breeding activities should be assessed in a more systematic manner to determine how these activities might be impacting larks during the breeding period. 9.6 Survey Other Suspected and Potential Lark Locations The detection of larks at Range 52 and the western portion of Range 50 confirmed our suspicions that larks might be breeding at other locations on JBLM. Improved access to safe locations on the Artillery Impact Area could ascertain whether our suspicions are correct. At a minimum, we recommend improved access to Range 50, Range 52, and Range 76 to accurately assess the number of pairs, confirm breeding, and survey these areas for banded birds. If access is not possible, the use of remote auditory devices could determine occupancy in these highly dangerous areas. In addition, suitable, but unoccupied, habitat in the Rainier Training Area (e.g., Weir and Johnson Prairies), 13th Division Prairie, and TA6 should continue to be surveyed for breeding larks.

SHLA_JBLM_2016_Final_2March_2017 Page 43

10.0 Acknowledgments CNLM is grateful to JBLM for their continued proactive approach, collaborative nature, and financial support for Streaked Horned Lark management and protection on the installation. In particular we thank Dave Clouse, Jim Lynch, Todd Zuchowski, Christa LeGrande, and Fiona Edwards with JBLM Fish and Wildlife. We thank the site managers and personnel at McChord and Gray Army Airfields for facilitating CNLM’s continued access to the airfields – Steele Clayton, Eileen Rodriguez, Zachary Miller, Steven Sawyer, and Annalee Thurber. Dr. Scott Pearson (WDFW) provided banding and monitoring materials, assisted with banding and technical review, as well as guidance and direction on many “larky” things. Dr. Randy Moore monitored the Corvallis airport population, and almost found a suitable match for an egg translocation. We also thank Christopher Lang with USDA for his continued cooperation. Several CNLM field biologists assisted with monitoring, and we thank Timothy Leque, Michael Warren, Monika Lapinski, Kerri Wheeler, Kelsi Potterf and Jason Derrick for their early commutes, and diligent efforts in the early mornings to determine breeding status, find nests, and track down band combinations. We also thank Dennis Vassar for providing access and ensuring our safety at Olympia Airport; and Brandon Palmer for facilitating access at Sanderson Airfield.

SHLA_JBLM_2016_Final_2March_2017 Page 44

11.0 Literature Cited

Anderson, J. K. 2010. Comparing endangered Streaked Horned Lark (Eremophila alpestris strigata) fecundity to other grassland birds. Masters Thesis. The Evergreen State College, Olympia, WA.

Beason, R. C. 1995. Horned Lark (Eremophila alpestris). In A. Poole and F. Gill, editors. The birds of North America, no. 195. The Academy of Natural Sciences, Philadelphia, Pennsylvania, USA; and The American Ornithologists' Union, Washington,D.C., USA.

Beyer, H. L. 2014. Geospatial Modelling Environment. Electronic resource [http://www.spatialecology.com/gme/]

Bibby, C. J., N. D. Burgess, and D. A. Hill. 1992. Bird census techniques. Academic Press, 257 pp.

Boulton, R.L., J.L. Lockwood, and M.J. Davis. 2009. Recovering small Cape Sable seaside sparrow (Ammodramus maritimus mirablilis) subpopulations: Breeding and dispersal of sparrows in the eastern Everglades 2008. Rutgers, The State University of New Jersey: Ecology, Evolution and Natural Resources.

Burnham, K. P., and D. R. Anderson. 1998. Model selection and inference: a practical information-theoretic approach. Springer-Verlag, New York.

Burnham, K. P., and D. R. Anderson. 2001. Kullnack-Leibler information as a basis for strong inference in ecological studies. Wildlife Research 28:111–119.

Camfield, A. F., S. F. Pearson, and K. Martin. 2010. Life history variation between high and low elevation subspecies of horned larks Eremophila spp. Journal of Avian Biology 41:273-281.

Camfield, A, F., S. F. Pearson, and K. Martin. 2011. A demographic model to evaluate population declines in the endangered Streaked Horned Lark. Avian Conservation and Ecology 6(2):4.

Cotton, P. A. 2003. Avian migration phenology and global climate change. Accessed electronically < www.pnas.orgcgidoi10.1073pnas.1930548100>

Dinsmore, S. J., G. C. White, and F. L. Knopf. 2002 Advanced techniques for modeling avian nest survival. Ecology 83: 3476-3488

Drovetski, S. V., S. F. Pearson, and S. Rohwer. 2005. Implications of mitochondrial DNA diversity on the conservation status of the streaked horned lark Eremophila alpestris strigata. Conservation Genetics 6:875-883.

Fletcher, K., D. Howarth, A. Kirby, R. Dunn, and A. Smith. 2013. Effect of climate change on breeding phenology, clutch size and chick survival of an upland bird. IBIS 155:456-463.

SHLA_JBLM_2016_Final_2March_2017 Page 45

Getz, W. M. and C. C. Wilmers. 2004. A local nearest-neighbor convex-hull construction of home ranges and utilization distributions. Ecography 27:489–505.

Gabrielson, G. W. 1985. Do not disturb nesting eiders! In Norsk Polarinstitutt Arbok 1984, 21-24

Harris, S., W. J. Cresswell, P. G. Forde, W. J. Trewhella, T. Woollard, and S. Wray. 1990. Home- range analysis using radio-tracking data – a review of problems and techniques particularly as applied to the study of mammals. Mammal Review, 20, 97–123.

Hejl, S. J., and J. Verner. 1990. Sequential versus initial observations in studies of avian foraging. Pages 166-173 in Avian foraging: theory, methodology, and applications (M. L. Morrison, C. Ralph, J Verner, and J. R. Jehl Jr., Eds.). Studies in Avian Biology, no. 13.

Keren, I.N., and S.F. Pearson. 2016. Streaked horned lark abundance and trends for the Puget lowlands and the lower Columbia River/Washington Coast, 2010-2015: Research Progress Report. Washington Department of Fish and Wildlife, Wildlife Science Division, Olympia, Washington.

Koenig, W. D. 1982. Ecological and social-factors affecting hatchability of eggs. Auk 99:526-536.

Koenig, W. D. 1988. Geographical ecology of clutch size variation in North American woodpeckers. Condor 88:499-504.

Kronland, B. J., K. Hill, and R. A. Martin. 2016. Prairie Habitat Management, Joint Base Lewis- McChord 2015 Annual Report. Unpublished report prepared for The Center for Natural Lands Management.

Martin, T. E. and G. R. Geupel. 1993. Nest-monitoring plots - methods for locating nests and monitoring success. Journal of Field Ornithology 64:507-519.

Mohr, C.O. 1947. Table of equivalent populations of North American small mammals. American Midland Naturalist 37:223–249.

Noble, G.K. 1939. The role of dominance in the social life of birds. Auk 56: 263–273.

Odum, E. P. and E. J. Kuenzler. 1955. Measurement of territory and home range size in birds. The Auk 72:128-137.

Pearson, S.F. 2003. Breeding Phenology, Nesting Success, Habitat Selection, and Census Methods for the Streaked Horned Lark in the Puget Lowlands of Washington. Natural Areas Program Report 2003-2. Washington Dept. of Natural Resources. Olympia, WA.

Pearson, S.F. and M. Hopey. 2007. Estimating streaked horned lark over-winter survival and site fidelity – Draft research progress report.doc

SHLA_JBLM_2016_Final_2March_2017 Page 46

Pearson, S. F., and M. Hopey. 2008. Identifying Streaked Horned Lark (Eremophila alpestris strigata) nest predators. Washington Department of Fish and Wildlife, Olympia, WA.

Pearson, S.F., and M. Hopey. 2005. Streaked Horned Lark Nest Success, Habitat Selection, and Habitat Enhancement Experiments for the Puget Lowlands, Coastal Washington and Columbia River Islands. Natural Areas Program Report 2005-1. Washington Dept. of Natural Resources. Olympia, WA.

Pearson, S.F., and M. Hopey. 2004. Streaked Horned Lark Inventory, Nesting Success and Habitat Selection in the Puget Lowlands of Washington. Natural Areas Program Report 2004- 1. Washington Dept. of Natural Resources. Olympia, WA.

Pearson, S.F., and S. M. Knapp. In prep. Considering spatial scale and reproductive consequences of habitat selection when managing grasslands for a threatened species. Washington Dept. of Fish and Wildlife.

Pearson, S. F., R. Moore, and S. M. Knapp. 2012. Nest exclosures do not improve Streaked Horned Lark nest success. J. Field. Ornithol. 83(3):315-322.

Pearson, S.F., M. Linders, I. Keren, H. Anderson, R. Moore, G. Slater, and A. Kreager. 2016. Survey protocols and strategies for assessing streaked horned lark site occupancy status, population abundance, and trends. Wildlife Science Division, Washington Department of Fish and Wildlife, Olympia, Washington.

Potterf, K. 2017. Using Thermochron iButtons to help monitor nest survival and increase daily survival rate accuracy for a threatened species, the Streaked Horned Lark (Eremophia alpestris strigata): A Pilot Study. Center for Natural Lands Management, South Sound Program, Olympia, WA.

Slater, G.L. 2016. Progress Report to ACUB: Post-fledging survival of streaked horned larks in prairie and airfield habitats in south Puget Sound: causes of mortality, habitat selection, and influence of habitat and body condition on survival. CNLM, Olympia, WA

Spencer, W. D. 2012. Home ranges and the value of spatial information. Journal of Mammology 93(4):929-947

Stinson, D. W. 2015. Draft Periodic status review for the Streaked Horned Lark in Washington. Washington Department of Fish and Wildlife, Olympia, Washington. 18 + iii pp.

Stinson, D. W. 2005. Washington State Status Report for the Mazama Pocket Gopher, Streaked Horned Lark, and Taylor’s Checkerspot. Washington Department of Fish and Wildlife, Olympia. 129+ xii pp.

Verner, J., D. Breese, and K. L. Purcell. 1998. Return rates of banded granivores in relation to band color and number of bands worn. Journal of Field Ornithology 71(1):117-125.

SHLA_JBLM_2016_Final_2March_2017 Page 47

Waite, T. A., Vucetich, J., Saurer, T., Kroninger, M., Vaughn, E., Field, K. & Ibargüen, S. 2005. Minimizing extinction risk through genetic rescue. Animal Biodiversity and Conservation 28.2: 121–130.

White, G. C., and K. P. Burnham. 1999. Program MARK: survival estimation from populations of marked animals. Bird Study 46:120–139.

Wolf, A. L. 2015. South Puget Sound Streaked Horned Lark (Eremophila alpestris strigata) Genetic Rescue Study – Report for Year 5. The Center for Natural Lands Management, South Sound Program, Olympia, Washington, October.

Wolf, A. L., and H. A. Anderson. 2014. Streaked Horned lark Habitat Management and Population Monitoring Report, Spring/Summer 2013. Center for Natural Lands Management, South Sound Program, Olympia, WA, April

Wolf, A. L., H. A. Anderson, and G. L. Slater. 2015. JBLM Larks 2014 Nest Monitoring Final Report. Annual report to JBLM. CNLM, Olympia, WA.

Wolf, A. L., H. A. Anderson, and G. L. Slater. 2016. JBLM Larks Monitoring Final Report 2015. Annual report to JBLM. Center for Natural Lands Management, South Sound Program, Olympia, WA, April.

SHLA_JBLM_2016_Final_2March_2017 Page 48

12.0 Appendices

Appendix 1. Streaked Horned Lark abundance transect survey results at five sites, JBLM, 2016

Site* Visit Date Males Females Unknown Age/Sex Total 5/20 12 1 2 15 13DIV 6/9 9 1 1 11 6/29 8 7 15 5/19 28 10 4 42 GAAF 6/7 28 7 7 42 6/20 27 3 7 37 5/18 7 2 9 MAFB 6/8 15 3 7 25 6/28 12 1 2 15 5/9 3 3 8 R50 5/23 9 6 15 6/21 5 3 8 5/3 8 3 13 R76 5/27 12 1 2 15 6/22 11 1 1 13 Survey 1 58 16 9 87 Total Survey 2 73 12 23 108 Survey 3 63 5 20 88 * 13DIV = 13th Division Prairie; GAAF = Gray Army Airfield; MAFB = McChord Airfield; R50 = Range 50, R76 = Range 76

SHLA_JBLM_2016_Final_2March_2017 Page 49

Appendix 2. Color-banded Streaked Horned Lark adults, South Puget Sound, 2016

Breeding Bands Original Band Sex Band Date USFWS# Natal Site* Location* Observed++ Combination++ 13DIV female RBt:GS RBt:GS 7/19/2013 2541-54359 GAAF 13DIV female BtG:WS BtG:WS 6/19/2013 2661-31209 13DIV 13DIV male YB:OS YB:OS 8/8/2014 2661-31274 MAFB 13DIV male GBk:WS GBk:WS 5/23/2014 2541-54377 13DIV 13DIV male GlY:MS+ GlY:MS+ 8/10/2011 2241-47833 13DIV 13DIV male S:Ph** BS:PhR 6/13/2009 2241-47866 13DIV 13DIV female R:MS R:MS 6/25/2011 2241-47897 13DIV 13DIV male YO:WS YO:WS(B) 7/24/2015 2711-89645 13DIV 13DIV male BkBt:WS BkBt:WS 6/8/2015 2711-89603 13DIV 13DIV male OBt:OS OBt:OS or OBtOS(B) 2015 unknown MAFB 13DIV male W:WS W:WS(B) 6/15/2015 2711-89616 13DIV 13DIV female BtB:WS BtB:WS(B) 6/18/2015 2711-89620 13DIV 13DIV female R:WS R:WS(B) 6/1/2015 2661-31298 13DIV GAAF male BkW:GS BkW:GS 7/19/2014 2661-31264 GAAF GAAF male BtB:GS BtB:GS 6/16/2014 2541-54395 GAAF GAAF male YB:GS YB:GS 8/8/2014 2661-31283 GAAF GAAF male B:GS B:GS 6/19/2013 2541-54317 GAAF GAAF female YBk:OS YBk:OS 6/13/2015 2711-83817 MAFB GAAF male OB:GS OB:GS 7/11/2014 2661-31244 GAAF GAAF male WBt:GS WBt:GS 6/1/2015 2661-31295 GAAF GAAF male OR:GS OR:GS 5/29/2014 2541-54392 GAAF GAAF male OY:OS OY:OS 7/24/2014 2661-31271 MAFB GAAF male OS:WPh OS:WPh 8/1/2014 2661-31258 MAFB GAAF male GY:GS GY:GS 7/11/2014 2661-31245 GAAF GAAF male BtR:GS BtR:GS 7/19/2013 2541-54360 GAAF GAAF female R:OS unknown Unknown unknown Unknown GAAF female RBk:GS RBk:GS(B) 7/20/2015 2711-89646 GAAF GAAF female RBt:GS RBt:GS(B) 7/13/2015 2711-89634 GAAF GAAF male Bk:GS Bk:GS or BkGS(B) 2015 unknown GAAF GAAF female WB:GS WB:GS 6/7/2015 2711-89601 GAAF GAAF male ?:GS unknown Unknown unknown GAAF GAAF male GB:GS GB:GS 6/11/2015 2711-89607 GAAF GAAF male WG:GS WG:GS 6/7/2015 2661-31300 GAAF GAAF female RG:GS RG:GS(B) 7/17/2015 2711-89641 GAAF GAAF male BB:GS BB:GS 5/20/2015 2661-31288 GAAF GAAF female B:GS B:GS(B) 6/13/2015 2711-83812 GAAF MAFB male YW:OS YW:OS 6/19/2015 2711-83818 MAFB MAFB male Bk:GS Bk:GS or BkGS(B) 2015 unknown GAAF MAFB male RW:OS RW:OS(B) 8/10/2015 2711-83874 MAFB MAFB male OB:GS OB:GS(B) 7/21/2015 2711-89651 GAAF MAFB male WB:OS WB:OS 7/2/2015 2711-83847 MAFB

SHLA_JBLM_2016_Final_2March_2017 Page 50

Breeding Bands Original Band Sex Band Date USFWS# Natal Site* Location* Observed++ Combination++ MAFB male WO:OS WO:OS 6/27/2015 2711-83840 MAFB MAFB male Bk:OS Bk:OS(B) 7/21/2015 2711-83855 MAFB MAFB male BtO:OS BtO:OS or BtO:OS(B) 2015 unknown MAFB MAFB male GY:OS GY:OS 7/22/2014 2661-31269 MAFB MAFB female R:OS R:OS(B) 7/5/2015 2711-83852 MAFB MAFB male BtBt:OS BtBt:OS or BtBt:OS(B) 2015 unknown MAFB MAFB female YO:GS YO:GS 6/11/2015 2711-89606 GAAF MAFB female YO:GS YO:GS(B) 7/24/2015 2711-89654 GAAF MAFB male Bt:OS Bt:OS or Bt:OS(B) 2014 or 2015 unknown MAFB MAFB male WR:OS WR:OS 6/27/2015 2711-83839 MAFB MAFB male YY:OS YY:OS 6/13/2015 2711-83816 MAFB MAFB female WG:OS WG:OS 7/2/2015 2711-83846 MAFB MAFB female BkY:OS BkY:OS 6/27/2015 2711-83836 MAFB MAFB male GW:OS GW:OS 6/22/2015 2711-83823 MAFB MAFB female BtBk:OS BtBk:OS(B) 7/29/2015 2711-83864 MAFB MAFB female OBk:OS OBk:OS 6/9/2015 2711-83810 MAFB MAFB female WW:GS WW:GS 6/10/2015 2711-89605 GAAF MAFB female BW:OS BW:OS 6/26/2015 2711-83835 MAFB MAFB male O:GS O:GS(B) 6/13/2015 2711-83809 GAAF MAFB male BtB:OS BtB:OS or BtB:OS(B) 2015 unknown MAFB MAFB male RW:GS RW:GS(B) 7/21/2015 2711-89647 GAAF OA male BtBt:GS BtBt:GS 8/6/2013 2541-54363 GAAF OA female R:GS R:GS(B) 6/11/2015 2711-89609 GAAF R76 male ?:BtS Unknown Unknown unknown Unknown R76 male YPh:WS YPh:WS 6/13/2014 2661-31217 13DIV R76 male BW:GS*** BW:GS*** 2014 unknown GAAF R76 female RW:GS RW:GS 7/1/2014 2661-31224 GAAF *Breeding location and Natal site codes: 13DIV = 13th Division Prairie, GAAF = Gray Army Airfield, OA = Olympia Airport; R76 = Range 76; **Male is likely hot pink/red – left, dark blue/USFWS - right, but the red and dark blue bands have since fallen off; + translocated from Corvallis OR.; *** duplicate band-combination applied in 2015; ++Color codes: B = dark blue, Bk = black, Bt = light blue, G = green, Gl = light green, O = orange, Ph = hot pink, R = red, S = USFWS silver, W = white, Y = yellow

SHLA_JBLM_2016_Final_2March_2017